docs(readme): carry the ToS caveat into the MANUAL install path too (README:218)

Reviewer found a third instance, and it is the one that most directly reproduces #136. README has two forms of the same LAN-mode install: the copy-paste AI prompt (line 132) and the handbook form (line 218). Line 180 explicitly asserts they are 'the same steps in handbook form'. They were not: line 132 (prompt) 'install OCP as a server so YOUR OWN DEVICES on the LAN can reach it (Claude Pro/Max are per-user accounts — review Anthropic's Usage Policy before extending access to other people)' + example keys: laptop, tablet line 218 (handbook) 'share with other devices on your network:' + 'create API keys for each PERSON/device' + no ToS pointer at all So a reader who takes the manual route instead of the copy-paste route is still walked into per-person key creation with zero ToS mention — reproducing #136's trigger through the door the first commit did not close. The caveat now matches its twin. Deliberately NOT scrubbing the wife-laptop / son-ipad example key names: they recur in seven further places, it is a bigger diff at a different severity, and it edges into scrubbing the maintainer's household out of their own documentation. With the pointer restored at 218 those examples inherit the caveat — which is exactly the posture § honest limits takes. It never forbids family sharing; it says it is the account holder's call and their risk, and refuses to hide that. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VqgWJcjxrjjL9L9SkpZyXR
docs(readme): fix the misdirected anchor + the same defect at line 52 (review fold-in)
2026-07-21 21:15:09 +00:00 · 2026-07-15 08:28:07 +10:00 · 2026-07-15 08:03:10 +10:00 · 2026-07-15 07:56:16 +10:00 · 2026-07-14 08:21:30 +10:00 · 2026-07-14 08:19:24 +10:00
18 changed files with 3279 additions and 83 deletions
@@ -24,7 +24,7 @@ One proxy. Multiple IDEs. All models. **$0 API cost.**

 There are several Claude proxy projects. OCP picks a specific lane: **align tightly with what `cli.js` actually does, observe + multiplex what's already there, don't extend the protocol.** What you get:

- **LAN multi-user keys** (v3.7.0) — share one Claude Pro/Max subscription with family, friends, or your own devices. Each user gets a per-key API token (no OAuth session leak), with independent usage tracking and one-line revocation.
+- **LAN multi-user keys** (v3.7.0) — reach one Claude Pro/Max subscription from your own devices across the LAN. Each device gets a per-key API token (no OAuth session leak), with independent usage tracking and one-line revocation. Pro/Max are **per-user** accounts — see [Sharing with family / a team — honest limits](#deployment-model--security-read-this) before extending access to other **people**.
 - **`ocp-connect` one-shot IDE setup** — one command on the client machine detects and configures Claude Code, Cursor, Cline, Continue.dev, OpenCode, and OpenClaw. No pasting `OPENAI_BASE_URL` six times.
 - **Response cache with per-key isolation + singleflight** (v3.13.0). Optional SHA-256 prompt cache, isolated per API key (cross-user pollution is impossible by hash construction, not by application logic), with stampede protection on concurrent identical prompts. Off by default. ([PR #65](https://github.com/dtzp555-max/ocp/pull/65), [PR #66](https://github.com/dtzp555-max/ocp/pull/66))
 - **Per-key request quotas** (v3.8.0). Daily / weekly / monthly limits per key — set a kid's iPad to 20/day, a partner's laptop to 100/week. ([PR #18](https://github.com/dtzp555-max/ocp/pull/18))
@@ -49,7 +49,7 @@ OCP and the alternatives serve adjacent but distinct needs. Pick the one that fi
 | GitHub stars / ecosystem size | small | large | mid |
 | Governance discipline (CI-enforced alignment with cli.js) | yes | n/a | n/a |

-**Plain English**: `claude-code-router` is the routing-and-switching power tool — pick it if you want to mix Anthropic, OpenAI, Gemini, and local models behind one endpoint. `anthropic-proxy` is the minimal forwarder. **OCP focuses on disciplined `cli.js`-aligned forwarding plus subscription multiplexing** — pick it if you want to share one Claude Pro/Max subscription across IDEs, devices, and people, with LAN auth, quotas, and a governance contract that prevents endpoint drift.
+**Plain English**: `claude-code-router` is the routing-and-switching power tool — pick it if you want to mix Anthropic, OpenAI, Gemini, and local models behind one endpoint. `anthropic-proxy` is the minimal forwarder. **OCP focuses on disciplined `cli.js`-aligned forwarding plus subscription multiplexing** — pick it if you want to reach one Claude Pro/Max subscription from your own IDEs and devices, with LAN auth, quotas, and a governance contract that prevents endpoint drift.

 ### Related: OLP — Open LLM Proxy

@@ -215,7 +215,7 @@ After install the `ocp` CLI lives at `~/ocp/ocp`. To put it on your PATH, either
 export OPENAI_BASE_URL=http://127.0.0.1:3456/v1
 ```

-**LAN mode** — share with other devices on your network:
+**LAN mode** — reach OCP from your own devices on the network (Claude Pro/Max are per-user accounts — see [Sharing with family / a team — honest limits](#deployment-model--security-read-this) before extending access to other people):
 ```bash
 # Enable LAN access with per-user auth (recommended)
 node setup.mjs --bind 0.0.0.0 --auth-mode multi
@@ -958,7 +958,13 @@ See [Subscription-pool (TUI) mode](#subscription-pool-tui-mode) and ADR 0007 PR-
 | `OCP_TUI_CWD` | `$HOME/.ocp-tui/work` | (TUI-mode) Scratch working directory where interactive claude sessions run. Transcripts land under `<HOME>/.claude/projects/<encoded-cwd>/`. Created automatically. |
 | `OCP_TUI_HOME` | *(auto)* | (TUI-mode) `HOME` claude runs under. **When unset, OCP picks it for you:** if `CLAUDE_CODE_OAUTH_TOKEN` is set → a **credential-isolated** scratch home `$HOME/.ocp-tui/home` (no `credentials.json`, env-token auth — **recommended**); if no env token → the operator's real home (legacy shared `credentials.json`). Setting this to an **explicit** path overrides the auto-default. The credential handling at that path still follows the env token: **with** the env token it is credential-free (env-token auth, no `credentials.json` written); **without** the env token (and the path ≠ real home) it uses the legacy symlinked-credentials scratch mode, which carries the credential-fork caveat — see ADR 0007. |
 | `OCP_TUI_ENTRYPOINT` | `cli` | (TUI-mode) Billing-classifier labeling: `cli` (default) pins `cc_entrypoint=cli` deterministically; `auto` lets claude self-classify via TTY detection; `off` leaves the inherited env untouched. Honest only when the spawn is a genuine interactive PTY — see ADR 0007. |
+| `OCP_TUI_EFFORT` | `low` | (TUI-mode) Effort level passed to the interactive `claude` as an explicit `--effort` flag: `low` (default), `medium`, `high`, `xhigh`, `max`, or `inherit` to omit the flag (the pre-flag behaviour: the pane inherits a HOME-dependent effort — the operator's `~/.claude/settings.json` `effortLevel` in real-home mode, claude's built-in default in env-token scratch mode). Explicit `low` cuts measured TTFT p50 by ~40% and collapses run-to-run variance ~15× versus an inherited `xhigh` (see `docs/plans/2026-07-13-tui-latency/`); proxied requests rarely benefit from extended thinking. Banner-verified to stay on the subscription pool (`· Claude Max`). An invalid value logs a warning and falls back to `low`. |
+| `OCP_TUI_STREAM` | `0` (off) | (TUI-mode) When `=1`, `stream:true` requests emit **real SSE `delta.content` chunks as `claude` generates them**, instead of buffering the turn and replaying it. Deltas come from `claude`'s own `MessageDisplay` hook (registered with `--settings` on the ordinary interactive spawn — banner-verified to stay on the subscription pool, `· Claude Max`). Granularity is **block-level**, not token-level. The transcript remains authoritative: the streamed text is asserted equal to it at end-of-turn, the auth-banner and truncation gates still run before anything is committed, and only the transcript text is cached. A turn whose stream cannot be reconciled with the transcript is **refused** (SSE error frame, not cached) and counted as `tui.streamDivergences` on `/health`. A total hook failure (e.g. `--settings` stops registering it after a `claude` version bump) is a *different, silent* failure mode — every streamed turn still succeeds, fully buffered, with no divergence and no error — so it is counted separately as `tui.streamZeroDeltaTurns` (streamed turns where the hook fired **zero** times) and logged as `tui_stream_zero_deltas`; watch it alongside `streamDivergences`. Default off — the buffered path is unchanged and remains the stable default. ⚠️ **Tool-using turns:** the transcript keeps only the model's **last** assistant message, so if the model narrates before calling a tool ("I'll check that file…") and that narration exceeds `OCP_TUI_STREAM_HOLDBACK`, it has already been streamed and cannot be retracted — the turn is then **refused** rather than served (measured live: Opus narrated 475 chars before a `Bash` call). If your deployment lets the model use tools (the TUI default, and anything with `OCP_TUI_FULL_TOOLS=1`), either raise `OCP_TUI_STREAM_HOLDBACK` above the typical narration length — the narration then stays held back and is correctly discarded, at the cost of a later first chunk — or leave streaming off. Streaming is best suited to tool-light chat proxying. See ADR 0007 (2026-07-13 amendment). |
+| `OCP_TUI_STREAM_HOLDBACK` | `100` | (TUI-mode, streaming) Characters withheld before the first chunk reaches the client. Two jobs. (1) It keeps the **auth-banner gate** alive under streaming, via a guarantee with two required halves: (i) nothing is emitted for a message until its trimmed accumulation exceeds 100 chars — past the default banner detector's reach, since real banners are ≤100 chars — and (ii) once a message boundary follows an emit, nothing further is ever emitted for the rest of the turn, and the turn is refused outright. Half (i) alone only covers a turn's first message; half (ii) is what covers an error banner rendered as a *later* message (e.g. after tool-using prose). Raise the holdback if you replace the detector via `CLAUDE_TUI_ERROR_PATTERNS` with patterns that can match longer messages — that only affects half (i); OCP warns at boot if you do. (2) It is the knob for **tool-using turns** — see the `OCP_TUI_STREAM` caveat below. Answers shorter than the holdback are simply delivered whole at end-of-turn, exactly as the buffered path does. |
+| `OCP_TUI_STREAM_DIR` | `$HOME/.ocp-tui/stream` | (TUI-mode, streaming) Directory holding the static `MessageDisplay` hook script + settings file, and the per-session delta sink (`<session-id>.jsonl`, removed at turn teardown). One sink **per session-id** — this is what keeps concurrent TUI turns (`OCP_TUI_MAX_CONCURRENT` ≥ 2) from interleaving one client's deltas into another's stream. |
+| `OCP_TUI_STREAM_POLL_MS` | `100` | (TUI-mode, streaming) Interval at which OCP drains the delta sink. The hook fires at block granularity (seconds apart), so a finer poll buys nothing. |
 | `OCP_TUI_MAX_CONCURRENT` | `2` | (TUI-mode) Max concurrent interactive TUI turns. **Independent** of `CLAUDE_MAX_CONCURRENT` (which bounds the `-p`/stream-json path; TUI never uses it). A TUI turn is heavy (per-request cold-boot of tmux+claude + up to `CLAUDE_TUI_WALLCLOCK_MS` wallclock), so the default is low to keep small hosts (e.g. a Pi 4) alive under a burst. Excess turns **queue** (bounded); a full queue yields a 503. See ADR 0007 PR-B amendment. |
+| `OCP_TUI_POOL_SIZE` | `0` (off) | (TUI-mode) Number of **pre-booted warm `claude` panes** kept ready, so a request does not pay the cold boot. `0` disables the pool entirely — the request path is then exactly the cold-boot path. Max `4`; an unparseable value disables it rather than guessing. **Measured on a Mac mini (Sonnet 4.6, `--effort low`): end-to-end p50 `10.17s` (n=6, pool off) → `6.00s` (n=12 warm hits) — −4.2 s / −41%** — the pool recovers both the ~1.2 s boot *and* ~2.9 s of post-input-bar init that a pane which has been idle a moment has already finished. **Cost:** each warm pane is a *live idle `claude` process* held whether or not a request ever arrives (peak processes ≈ pool size + `OCP_TUI_MAX_CONCURRENT` + 1 booting replacement) — which is why it is opt-in. Panes are **single-use**: one turn, then killed and replaced in the background. The **first request after start (and after any model switch) is always a cold miss** — the pool warms the most recently requested model, since OCP cannot know which model the next caller wants. See `docs/plans/2026-07-13-tui-latency/`. |
 | `OCP_SKIP_AUTH_TEST` | *(unset)* | When `=1`, skip the `claude -p` auth probe during `setup.mjs`. After 2026-06-15 this probe draws from the Agent SDK credit pool; set this to avoid burning a metered credit on re-installs or `ocp update` runs. Auth is validated at the first real request. |
 | `OCP_TUI_FULL_TOOLS` | *(unset)* | (TUI-mode, **single-user only**) When `=1`, grant the interactive session the **same tool surface as the `-p` path** — `--allowedTools` (+ optional `--mcp-config`, read from `CLAUDE_ALLOWED_TOOLS` / `CLAUDE_MCP_CONFIG`) — instead of the default MCP-walled, built-in-tools-only set. Lets a trusted single-operator TUI deployment run a **tool-using / MCP agent** (e.g. an OpenClaw assistant) on the subscription pool. Safe because TUI **refuses to boot under `AUTH_MODE=multi`** (hard exit) — no guest key can ever reach the TUI path, so this gate cannot expose tools to an untrusted caller. (Under `AUTH_MODE=shared` + `OCP_TUI_ALLOW_LAN=1`, anyone holding the single shared key reaches it — that is the existing TUI trust model, unchanged.) Note: `--dangerously-skip-permissions` / `CLAUDE_SKIP_PERMISSIONS` is **not** supported for TUI — claude v2.1.x shows an interactive bypass-acceptance screen in headless tmux that cannot be answered, bricking the pane. Use scratch-home `settings.json` `additionalDirectories` instead. See [Subscription-pool (TUI) mode](#subscription-pool-tui-mode) and ADR 0007. |

@@ -1032,13 +1038,43 @@ Then restart OCP. At boot you will see (with the env token set, isolated home au
 ### What changes / what doesn't

 - **Callers see no API change.** The response is a normal OpenAI completion object or chunked SSE — identical wire format.
- **No real token streaming.** TUI-mode buffers the full response then replays it as chunked SSE. You will see a delay then the complete response rather than real-time tokens.
+- **Real streaming is opt-in (`OCP_TUI_STREAM=1`), and off by default.** By default TUI-mode buffers the full response and replays it as chunked SSE — you see a delay, then the complete response. Set `OCP_TUI_STREAM=1` and `stream:true` turns emit real SSE `delta.content` chunks as `claude` renders them, sourced from `claude`'s own `MessageDisplay` hook (byte-faithful raw markdown, on the subscription pool, no `-p`). Two honest caveats: granularity is **block-level** — the hook fires once per rendered block, so a handful of chunks per answer, scaling with length, not token-by-token; and it moves the **first** byte, not the last, so a consumer that must parse a complete reply gains nothing. The transcript stays authoritative: every streamed turn is asserted against it at the end, and a turn whose stream disagrees is **failed rather than served** (watch `tui.streamDivergences` on `/health`). Evidence: [`docs/plans/2026-07-13-tui-latency/streaming-spike.md`](docs/plans/2026-07-13-tui-latency/streaming-spike.md).
 - **Cache and singleflight work normally.** TUI-mode writes the buffered response to the cache on success; cache-hits skip the interactive turn entirely.
 - **The host's `CLAUDE.md` / auto-memory is never injected.** OCP is a proxy — the proxied client (OpenClaw / your IDE) owns its own context and memory. TUI-mode always runs `claude` with `CLAUDE_CODE_DISABLE_CLAUDE_MDS` + `CLAUDE_CODE_DISABLE_AUTO_MEMORY`, so a `CLAUDE.md` on the OCP host can never leak into proxied turns (verified live; see #4). Built-in tool schemas + the interactive system prompt remain (the inherent ~20–35K context floor of interactive mode); MCP is hard-disabled.
 - **Authenticate via `CLAUDE_CODE_OAUTH_TOKEN` in a credential-isolated home (recommended).** tmux does not forward the parent process's env to the pane, so OCP sets the token explicitly on the spawned `claude` when `CLAUDE_CODE_OAUTH_TOKEN` is present. But passing the token is **not enough on its own**: interactive `claude` *prefers* `~/.claude/.credentials.json` over the env var (unlike the `-p` path), so a stale `credentials.json` would shadow the token. With the env token set and `OCP_TUI_HOME` unset, OCP therefore runs claude in a **credential-isolated home** (`$HOME/.ocp-tui/home`) that has **no `credentials.json`** — so the env token is the only credential and is authoritative, and claude never runs the token-refresh path (so the single-use refresh token can't be corrupted by the spawn/teardown cycle). On a long-running host the credentials.json path produced a permanent `Please run /login · API Error: 401` that re-login could not fix (the next spawn re-corrupted it); the isolated home ends that at the root. Transcripts land under the same isolated home, so the answer-reader is unaffected. Without the env token, claude falls back to the real home's `credentials.json` (byte-for-byte the previous behaviour). (The token is visible in `ps` on the pane command — acceptable for the single-user A-path; the multi-user B-path is refused at boot.) See ADR 0007 PR-C / PR-D amendments.
 - **Stale tmux sessions are reaped.** The pane's `claude` is a child of the tmux server (not OCP), so OCP cannot reap it directly; `claude` zombies can otherwise accumulate as `<defunct>` over a long-running host. OCP reaps them at boot and on a 15-min idle sweep by issuing `tmux kill-server` — but **only when no foreign tmux session remains** (it never disrupts a co-hosted `olp-tui-*` instance). See ADR 0007 PR-C amendment.
 - **Default path unchanged.** Unset `CLAUDE_TUI_MODE` and restart → `callClaude` / `callClaudeStreaming` are used again, byte-for-byte identical to today.
 - **Concurrency is bounded separately.** TUI turns are heavy (per-request cold-boot + long wallclock), so the TUI path has its own limiter — `OCP_TUI_MAX_CONCURRENT` (default `2`), independent of `CLAUDE_MAX_CONCURRENT`. Excess turns queue; a full queue returns a 503. Tune it up only on a host that can run more interactive `claude` sessions at once.
+- **Optional warm pane pool (`OCP_TUI_POOL_SIZE`, default off).** Pre-boots panes so a request skips the cold boot — measured p50 `10.17s` → `6.00s` (−41%). Pooled panes are **single-use** (one turn, then killed and replaced in the background), each carrying its own fresh `--session-id`, so one session still means one exchange and no earlier-turn text can leak into a later answer. They are named `ocp-tui-<port>-p<hex>` and coexist with the reaper by design: the sweep **drains the pool first**, then reaps (so `kill-server` still flushes `<defunct>` zombies), then the pool refills in the background. Drain→reap→resume is synchronous, so no request can land mid-sweep; a request arriving while the pool is still re-booting simply misses it and cold-boots. A live pooled pane is never reaped — **including one that is still booting**, whose tmux session already exists — while an *orphaned* one (left by a previous process generation) still is.
+
+### ⚠️ Latency: TUI mode has a ~6-second floor, and it is immovable
+
+**TUI mode cannot serve real-time or interactive-latency consumers.** This is a hard property of the
+path, stated plainly so you can rule it out before building on it:
+
+| | measured |
+|---|---|
+| **TTFT floor (first token)** | **≈ 6 s** — immovable |
+| cold boot → input bar ready | ~1 s (per request; not the bottleneck) |
+| OCP's own overhead above the CLI | ~4 s (n=1 same-turn decomposition) |
+| direct Anthropic API, same prompt (for scale) | 0.84–1.64 s |
+
+The ~6 s floor is the `claude` CLI itself: it always injects the full Claude Code system prompt plus
+its tool definitions before your prompt, on every turn, no matter what you ask. No flag removes it
+(`--exclude-dynamic-system-prompt-sections` was measured: **no effect** on the floor). Extended
+thinking is *not* the cause — `OCP_TUI_EFFORT` already defaults to `low`, which is what cuts a
+formerly-inherited `xhigh` down to this floor and collapses its variance.
+
+On top of the floor you pay the model's generation time (a function of output length). Progressive
+output is not wired up **yet** (see "No real token streaming" above — it is achievable and planned),
+so today a turn returns as one blob once generation completes. Note that streaming, when it lands,
+will move the *first* byte earlier — it does **not** shorten the turn, and a consumer that needs the
+complete answer gains nothing from it.
+
+**Use TUI mode for**: batch, background, and latency-insensitive work where the subscription pool is
+the point. **Do not use it for**: anything a person is waiting on interactively, or any consumer with
+a sub-5-second budget. Full measurements and methodology:
+[`docs/plans/2026-07-13-tui-latency/`](docs/plans/2026-07-13-tui-latency/).

 ### Monitoring drift via `/health`

@@ -1052,12 +1088,26 @@ Then restart OCP. At boot you will see (with the env token set, isolated home au
  "entrypointMismatches": 0,   // count of cli-expected-but-got-other turns — ALERT if this climbs
  "inflight": 1,               // TUI turns running right now
  "queued": 0,                 // TUI turns waiting for a concurrency slot
-  "maxConcurrent": 2           // OCP_TUI_MAX_CONCURRENT
+  "maxConcurrent": 2,          // OCP_TUI_MAX_CONCURRENT
+  "pool": {                    // warm pane pool — null when OCP_TUI_POOL_SIZE=0 (the default)
+    "size": 2,                 // target warm panes (OCP_TUI_POOL_SIZE)
+    "warm": 2,                 // panes ready right now — each is a LIVE idle claude process
+    "booting": 0,              // replacement panes currently pre-booting
+    "model": "claude-sonnet-4-6", // the model being warmed (the most recently requested one)
+    "hits": 12,                // requests served by a warm pane
+    "misses": 1,               // requests that fell back to the cold boot (the 1st is always one)
+    "boots": 14,               // panes successfully pre-booted
+    "bootFailures": 0,         // pre-boots that genuinely never reached the input bar — WATCH this
+    "cancelled": 4,            // in-flight boots OCP killed on purpose (drain / model switch) — not faults
+    "dropped": 8               // panes discarded unused (drain sweep / expired / unhealthy)
+  }
 }
 ```

 Alert on `entrypointMismatches > 0` (or `lastEntrypoint !== "cli"`): it means a turn drew from the metered Agent SDK pool instead of the subscription. `inflight` / `queued` show how close the TUI path is to its concurrency cap.

+With the pool on, `hits` / `misses` is the hit rate (a steady single-model consumer should sit near 100% after the first request), and `warm` is your standing idle-process cost. A climbing `bootFailures` means panes are not reaching their input bar — the pool then degrades safely to the cold path, but latency reverts to the un-pooled numbers. `cancelled` counts boots OCP killed *on purpose* (a drain, a model switch) and is **not** a fault signal — do not alert on it. A steadily climbing `dropped` is likewise normal: the 15-min reap sweep drains and re-boots the pool on every tick so `kill-server` can still flush `<defunct>` zombies.
+
 ### Kill-switch

 ```bash
@@ -56,7 +56,7 @@ Add `CLAUDE_TUI_MODE=true` as an opt-in flag in `server.mjs`.
 3. The serialized prompt (from `messagesToPrompt`) is pasted via `tmux send-keys … "$(cat file)"` + a separate `Enter` key event.
 4. The answer is read from claude's native JSONL transcript at `<HOME>/.claude/projects/<encoded-cwd>/<session-id>.jsonl`, polling until a `turn_duration` system event or the wall-clock cap (`CLAUDE_TUI_WALLCLOCK_MS`, default 120 s).
 5. The string answer is returned to OCP's existing downstream (singleflight → cache write-back → `completionResponse` / `streamStringAsSSE`) — **same contract as `callClaude`**.
-6. Streaming requests are buffered then replayed as chunked SSE (no real token streaming — deliberate; "don't build fragile features").
+6. Streaming requests are buffered then replayed as chunked SSE (no real token streaming — deliberate; "don't build fragile features"). **Superseded for `stream:true` when `OCP_TUI_STREAM=1` — see the 2026-07-13 amendment below. The buffered path remains the default and is unchanged.**

 ### Billing-classifier labeling (`OCP_TUI_ENTRYPOINT`, PR-4)

@@ -333,6 +333,61 @@ The original "Home strategy" section and PR-C's `prepareTuiHome` comment warned

 ---

+## Amendment (2026-07-13) — real SSE streaming via the `MessageDisplay` hook (`OCP_TUI_STREAM`)
+
+**Supersedes**: Request-flow step 6 above ("no real token streaming — deliberate"), for `stream:true`
+requests when `OCP_TUI_STREAM=1`. The buffered path stays the default and is byte-for-byte unchanged.
+
+**Context.** Step 6 was written when the interactive CLI appeared to expose no byte-faithful
+incremental source. A prereq spike (`docs/plans/2026-07-13-tui-latency/streaming-spike.md`) confirmed
+three obvious sources are dead ends — the transcript JSONL grows one *whole event* at a time (the
+answer lands as a single line ~0.3 s before the terminal marker); `tmux capture-pane` yields a
+*rendered* view whose markdown source is unrecoverable (an H2 and a bold span produce identical ANSI);
+`--debug-file` logs stream *timing*, never stream *content*. Every interface that does emit
+`text_delta` (`--output-format stream-json`) requires `-p`, which moves the request to the **metered**
+`sdk-cli` pool — precisely what TUI-mode exists to avoid.
+
+**Decision.** Consume `claude`'s own **`MessageDisplay`** hook, registered via `--settings` on the
+ordinary interactive spawn (no `-p`, no `--bare`). Each fire delivers the **raw markdown source** of an
+incremental `delta` on the hook's stdin. Verified live (claude 2.1.207, sonnet-4-6): banner stays
+`· Claude Max` and the transcript `entrypoint` stays `cli` (subscription pool); `concat(deltas) === T`
+byte-exactly; `T.startsWith(concat(deltas[0..n]))` at every *n*. This is **forwarding, not inventing**
+— ALIGNMENT.md **Class B**. No `cli.js` citation applies: the TUI spawn is OCP-owned surface (this
+ADR), the hook payload is claude's own published contract, and the SSE wire shapes are the OpenAI
+chat/completions streaming spec adopted by **ADR 0006** (the emitters are literally the `-p` path's).
+
+**The transcript remains authoritative.** It is still the terminal-turn signal, still the source of the
+returned/cached text `T`, and still the input to the honesty gates (auth-banner detection C-1,
+`truncated` C-2). The delta stream is a low-latency **mirror**, never a replacement. At end of turn OCP
+asserts the streamed bytes against `T`: equal → serve; a strict *prefix* of `T` → top up from the
+transcript (client still receives exactly `T`); **not** a prefix → **refuse the turn** (SSE error frame,
+no cache, `tui.streamDivergences++`). Serving text the transcript disagrees with is the failure class
+ALIGNMENT.md exists to prevent, so streaming fails loud rather than degrading quietly.
+
+**Consequences / constraints recorded for future authors:**
+
+- **Opt-in, default OFF.** The buffered path is stable production; streaming does not change it.
+- **Per-`session_id` sink is mandatory, not an optimization.** `OCP_TUI_MAX_CONCURRENT` defaults to
+  **2** — two `claude` panes already run concurrently. A single shared sink would interleave one
+  client's deltas into another's stream. The hook writes to `<dir>/<session_id>.jsonl`, the path
+  delivered through the *pane's own env* (`OCP_TUI_STREAM_FILE`); OCP reads only its own turn's file.
+  Verified with two concurrent streamed turns (ALPHA/BRAVO): zero cross-contamination.
+- **Warm-pool compatible (a separate in-flight PR depends on this).** The hook script and the settings
+  file are **static** — nothing request-specific is baked in at spawn time. The sink path derives from
+  the session-id, which for a pre-booted pane is fixed at boot.
+- **The hook is synchronous** (`forceSyncExecution: true` — `claude` *blocks* on it). The hook script
+  must write and exit; it does one `cat` append and nothing else. Measured: p50 **7.2 ms** per fire,
+  ~50 ms across a whole turn — noise against a 6–10 s turn. Do not add work to it.
+- **Thinking blocks do not fire the hook** — verified on a substantive Opus/`xhigh` reasoning turn (see
+  the PR evidence), not merely inferred from the `final:true` call site. This must be **re-verified** if
+  the hook is ever pointed at a new model/effort tier: a thinking delta reaching a client would be
+  unretractable, and the `concat === T` assertion can only *detect* that after the fact, never prevent
+  it. The first-bytes **holdback** (`OCP_TUI_STREAM_HOLDBACK`, default 100 chars) is the same
+  prevention-not-detection reasoning applied to the auth-banner gate.
+- **Block-level granularity**, scaling with answer length — not token-level. Do not promise otherwise.
+- **It moves the first byte, not the last.** Only a progressively-rendering consumer benefits; it does
+  not move TUI-mode's ~6 s TTFT floor.
+
 ## Provenance

 TUI-mode originated in a prototype contributed via PR #101 (see the PR for author attribution). The productionization design is in `docs/superpowers/specs/2026-05-30-tui-mode-production-design.md`. Spikes S1–S6 / T1–T6 were validated live on the test host against `claude v2.1.158`.
@@ -0,0 +1,208 @@
+# ADR 0008 — TUI Warm Pane Pool
+
+**Date:** 2026-07-13
+**Status:** Proposed
+**Extends:** [ADR 0007](0007-tui-interactive-mode.md) (TUI interactive mode). This ADR does not
+change ADR 0007's billing-pool argument, security posture, or kill-switch — it adds a latency
+optimization *inside* the TUI spawn machinery ADR 0007 owns.
+
+---
+
+## Context
+
+TUI mode (ADR 0007) serves every request by cold-booting a fresh `tmux` session running an
+interactive `claude`, submitting one prompt, reading the native transcript, and killing the
+session. That cold boot is paid on **every** request.
+
+[`docs/plans/2026-07-13-tui-latency/`](../plans/2026-07-13-tui-latency/README.md) measured the
+TUI path and listed a warm pane pool as backlog item #3, costed at "**~1.0 s**" (the observed
+boot-to-input-bar time). Instrumenting the real request path showed that estimate is **~4×
+too low**. Phase decomposition of the cold path (n=6 medians, Sonnet 4.6, `--effort low`,
+through a real OCP instance):
+
+| Phase | Median |
+|---|---|
+| prep (trust cwd, write prompt file) | 2 ms |
+| `tmux new-session` | 27 ms |
+| **boot → input bar ready** | **1232 ms** |
+| paste (`load-buffer` + `paste-buffer`) | 8 ms |
+| paste-verify poll | 426 ms |
+| **submit → transcript terminal** | **8458 ms** |
+| teardown | 8 ms |
+| **total** | **10162 ms** |
+| *claude's own reported `turn_duration`* | *5539 ms* |
+| **OCP-side overhead** | **4490 ms** |
+
+The `submit → terminal` phase exceeds claude's own `turn_duration` by **~2.9 s**. That gap is
+**post-input-bar initialization inside `claude`** — work that a pane which has merely *sat idle
+for a few seconds* has already completed. A direct spike confirmed it: an identical pane, idle
+12 s before receiving the same prompt, completed its turn in a median 5537 ms versus 7980 ms
+cold.
+
+So a warm pane recovers **~1.26 s of boot *and* ~2.9 s of in-`claude` cold start** — not the
+~1.0 s the plan predicted.
+
+The reason this was worth a pool rather than a "keep one session and reuse it" cache is a
+hazard already flagged in the code. `lib/tui/transcript.mjs` returns the **last text-bearing
+assistant entry in the whole transcript file**, which is correct *only* under OCP's
+one-session-per-request model, and it says so:
+
+> *"If a future warm-pool ever reuses a session WITHOUT a fresh session-id / clear, earlier-turn
+> text could leak — that author must add user-line scoping here."*
+
+Reusing a pane for a second turn puts two exchanges in one transcript and would leak the earlier
+turn's text into the later turn's answer — a **cross-request data leak**, not merely a bug.
+
+---
+
+## Decision
+
+Add an **opt-in pool of pre-booted, single-use `claude` panes**, `OCP_TUI_POOL_SIZE` (default
+`0` = off, max `4`). Implementation: `lib/tui/pool.mjs`.
+
+### 1. Panes are SINGLE-USE. This is the load-bearing rule.
+
+A pooled pane serves **exactly one turn**, then is killed and replaced in the background. Each
+pane is booted with its **own fresh `--session-id`**, fixed at spawn, and the turn locates its
+transcript by that id.
+
+This preserves one-session-per-request exactly, so the `transcript.mjs` hazard above **does not
+arise** and no user-line scoping was needed. The warning in `transcript.mjs` is deliberately
+left standing, now annotated: it still binds anyone who later wants a pane to serve a second
+turn, or to reset a session with `/clear` and reuse it. **Neither is permitted without first
+adding user-line scoping to the transcript reader.**
+
+Rejected alternative — *reuse a pane for N turns, `/clear` between* — is strictly cheaper
+(no re-boot per request) and was rejected on exactly this basis. The latency win is not worth a
+cross-request text-leak surface guarded only by a `/clear` that we cannot verify landed.
+
+### 2. The pool is keyed by model, and a MISS is always safe.
+
+`--model` is fixed at spawn, so a pane can only serve the model it booted with. A pool miss
+falls back to the existing cold-boot path with **zero behavioural difference**. There is no
+boot-time pre-warm and no configured model: OCP cannot know which model the next caller wants,
+so the pool warms the **most recently requested** model. Consequence, stated plainly: **the
+first request after start, and the first after any model switch, is always a cold miss.**
+
+### 3. The pool and the session reaper coexist by an explicit invariant.
+
+This is the subtle part. `reapStaleTuiSessions()` kills every session matching this instance's
+`ocp-tui-<port>-` prefix, and issues `tmux kill-server` when no foreign session remains (the
+only mechanism that can reap `<defunct>` `claude` zombies — the pane's `claude` is a child of
+the tmux *server*, not of node). A warm pooled pane **is** one of our own sessions, alive and
+idle **by design** — and the periodic sweep runs precisely **when the instance is idle**, i.e.
+exactly when the pool is full.
+
+The invariant, stated in a comment above `reapStaleTuiSessions` and pinned by tests:
+
+1. **A live pooled pane is never reaped — including one that is still BOOTING.** The reaper
+   takes a `spare` set of **exact session names** supplied by the pool's live registry.
+2. **An orphaned pooled pane IS still reaped.** Membership is by **exact name from a live
+   in-memory registry, never by name shape**. A pane the pool no longer owns — handed out,
+   dropped, cancelled, or left behind by a previous process generation (whose registry died with
+   it) — is absent from `spare` and is killed like any other stale session. **Fail-safe:
+   omitting `spare` reaps *more*, never less.** Pool panes are named `ocp-tui-<port>-p<hex>`
+   purely for operator legibility; that shape is *not* the exemption mechanism.
+3. **`kill-server` is suppressed while any pane is spared** (it would kill a live child of the
+   tmux server). Therefore **the pool is DRAINED immediately before every sweep**, so `spare` is
+   empty on the normal tick and `kill-server` still fires. Without the drain, a permanently-full
+   pool would **permanently disable zombie reaping** — the pool would silently break the thing
+   the sweep exists to do. The drain costs one pane re-boot per tick (15 min).
+
+The `spare` mechanism is belt-and-braces given the drain: it makes it impossible for a reap call
+site that *forgets* to drain to kill a live pane.
+
+### 4. The pool tracks its in-flight boot BY NAME, not as a count.
+
+`bootTuiPane` creates the tmux session **synchronously** and only *then* waits (up to
+`POOL_BOOT_MS`, 20 s) for the input bar. So **a pooled tmux session can be live for ~20 s before
+its boot resolves.** A pool that tracked in-flight boots as a *count* could not name that
+session, and this produced two real bugs (both caught in review, both now regression-tested):
+
+- the periodic sweep **killed the booting pane** (it could not be spared), then left the pool
+  empty with nothing scheduled, and logged the exact `tui_pool_boot_failed` warning operators are
+  told to alert on — for a completely healthy drain;
+- graceful shutdown **orphaned a live, authenticated, idle `claude`**: `gracefulShutdown` calls
+  `process.exit(0)` in the same tick as the drain (TUI panes are tmux children, so node's
+  `activeProcesses` set is empty and the "wait for children" path exits immediately), so any
+  cleanup deferred to a `.then()` never ran.
+
+The pool therefore **mints each pane's identity up front** (`{sessionId, name}`, name derived
+from the session-id so `tmux ls` correlates to the transcript file) and holds it in
+`_bootingPane`. `liveNames()` includes it; `drain()` kills it **synchronously**. A generation
+counter distinguishes *"cancelled by us"* from *"genuinely failed"*, so a drain never inflates
+`bootFailures` and `resume()` reliably starts a fresh boot.
+
+### 5. Refills take no concurrency slot, and are serialized.
+
+A refill boot deliberately does **not** take a `TuiSemaphore` slot: those slots bound concurrent
+*turns* and belong to real requests, and charging a background pre-boot against them would let
+the pool starve the traffic it exists to speed up. It cannot leak a slot either, since it never
+holds one. Boots are **serialized** (one at a time): two cold boots racing an in-flight turn were
+observed to overrun even the generous pool readiness cap. A genuinely failed boot does **not**
+re-kick the chain (backoff — a broken `claude` must not respawn forever).
+
+Background boots get a more generous readiness cap (`POOL_BOOT_MS` = 5 × `BOOT_MS`): `BOOT_MS` is
+tight because a *client* is blocked on it, which is not true of a pre-boot. Slow ≠ broken.
+
+---
+
+## Consequences
+
+### Cost — standing processes, paid whether or not a request arrives
+
+**A warm pane is a live idle `claude` process.** Peak process count is
+`OCP_TUI_POOL_SIZE` + `OCP_TUI_MAX_CONCURRENT` + 1 (booting replacement). This is the whole
+reason the pool is **default-off**: an operator must opt into holding processes for traffic that
+may never come. Size is clamped to `POOL_MAX_SIZE` = 4; an unparseable value **disables** the
+pool rather than guessing.
+
+Panes carry a 10-minute TTL and are health-checked at hand-out; a dead or degraded pane becomes
+a **miss** (cold path), never a hung turn.
+
+### Benefit
+
+Measured end-to-end through a real OCP instance (Sonnet 4.6, `--effort low`):
+**p50 10.17 s (n=6, pool off) → 6.00 s (n=12 warm hits) — −4.2 s / −41%.**
+
+### The floor is unchanged
+
+The pool does not touch the **~6 s TTFT floor** documented in the latency plan (claude always
+prefills the full Claude Code system prompt). TUI mode remains unsuitable for interactive /
+real-time consumers; it is for batch and background work. This ADR does not change that
+conclusion.
+
+### Observability
+
+`/health`'s `tui` block gains a `pool` sub-object (`null` when off): `size`, `warm`, `booting`,
+`model`, `hits`, `misses`, `boots`, `bootFailures`, `cancelled`, `dropped`. A climbing
+`bootFailures` means panes are not reaching their input bar — the pool then degrades safely to
+the cold path, but latency reverts to the un-pooled numbers. A steadily climbing `dropped` is
+**normal** (the 15-min sweep drains and re-boots the pool on every tick, by design — see
+Decision 3).
+
+### ALIGNMENT authorization
+
+- **Class B / OCP-owned.** The warm pool is process management around the `claude` CLI — the
+  same category as the existing tmux session lifecycle and the defunct-session reaper it extends.
+  **`cli.js` does not perform this operation, and no `cli.js` citation applies**; the authority
+  is ADR 0007 (which owns the TUI spawn machinery) plus this ADR. This is `ALIGNMENT.md` Rule 2's
+  Class B citation requirement, discharged explicitly rather than by silence.
+- **The `/health` extension** adds sub-fields to the `tui` block. That block is **owned by ADR
+  0007** and post-dates ADR 0006's v3.16.4 grandfather snapshot, so it is not part of the frozen
+  B.2 inventory. The change is additive — every pre-existing `/health` field keeps a
+  byte-identical value, and `pool` is `null` unless the operator opts in — which is the
+  behaviour-preserving bar ADR 0006 sets. This ADR records that authorization.
+- **No spawn argument changed.** `buildTuiCmd` is byte-identical; the pool calls it with the same
+  arguments. Banner-verified on live pooled panes: `· Claude Max`, never `API Usage Billing`
+  (the `--bare` trap documented in the latency plan).
+
+### What a future contributor must not undo
+
+- **Do not let a pane serve a second turn** (or `/clear`-and-reuse one) without first adding
+  user-line scoping to `lib/tui/transcript.mjs`. That is a cross-request text leak, not a perf
+  tweak. See Decision 1.
+- **Do not remove the drain-before-sweep.** It is what keeps `kill-server` zombie reaping alive.
+  See Decision 3.
+- **Do not go back to counting in-flight boots.** The pool must be able to *name* a session that
+  exists but has not finished booting. See Decision 4.
@@ -23,6 +23,8 @@ New ADRs increment from the highest existing number. Filenames are
 | [0004](0004-openclaw-auto-sync.md) | OpenClaw Auto-Sync | Why `scripts/sync-openclaw.mjs` runs on `ocp update`, what its scope boundary is (writes only `models.providers["claude-local"].models` and `agents.defaults.models["claude-local/*"]`), and the idempotency contract. |
 | [0005](0005-no-multi-provider.md) | No Multi-Provider | Why OCP stays single-provider (Anthropic-via-cli.js) and does not extend to OpenAI / Gemini / OpenRouter. Cost estimate: ~7 weeks for a v1 that buys neither moat nor commercial readiness. Separate commercial work starts in a separate repo. |
 | [0006](0006-openai-shim-scope.md) | OpenAI Shim Scope | The Class A / Class B taxonomy. Class A endpoints (`cli.js`-mirror) keep Rules 1–5 verbatim; Class B endpoints (OCP-owned compatibility surface — `/v1/chat/completions`, `/v1/models`, admin endpoints) are anchored to OpenAI's spec (B.1) or to an authorizing ADR (B.2). Triggered by PR #99 (external `response_format` honoring). Grandfathers the existing B.2 inventory at v3.16.4. |
+| [0007](0007-tui-interactive-mode.md) | TUI Interactive Mode | Why TUI-mode spawns an interactive `claude` in a tmux pane (no `-p`) to reach the **subscription** billing pool (`cc_entrypoint=cli`) rather than the metered Agent SDK pool. Owns the TUI spawn machinery: entrypoint labeling, credential-isolated home, MCP hard-disable, session namespace + defunct-session reaping, the independent concurrency bound, and the `/health` `tui` block. **Single-user only** — hard FATAL on multi-user configs. |
+| [0008](0008-tui-warm-pane-pool.md) | TUI Warm Pane Pool | Why `OCP_TUI_POOL_SIZE` pre-boots **single-use** `claude` panes (one turn each, own `--session-id`) — and why reuse is forbidden (`transcript.mjs` returns the last assistant entry in the file, so a reused session leaks the earlier turn's text). Measured −41% end-to-end. Defines the pool↔reaper invariant (exemption by exact name from a live registry; drain before every sweep so `kill-server` zombie reaping survives) and the standing idle-process cost. Extends ADR 0007. |

 ## When to write a new ADR

@@ -0,0 +1,236 @@
+# TUI-mode latency: measured floor, and the four things worth fixing
+
+**Date**: 2026-07-13
+**Status**: findings + backlog. **Superseded in part** — see the dated update boxes below.
+Item #1 shipped ([#156](https://github.com/dtzp555-max/ocp/pull/156)); item #2 is **dead**
+([`streaming-spike.md`](streaming-spike.md)); item #4 measured, **no effect**; item #3 stands.
+**Measured on**: Mac mini / macOS 26.5.2 / Claude Code **v2.1.207** / Sonnet 5 / Claude Max subscription / **real-home mode** (no `CLAUDE_CODE_OAUTH_TOKEN`, no `OCP_TUI_HOME` in the service env)
+**Evidence**: [`measurements.jsonl`](measurements.jsonl) — **n=15** (3 configs × 5) · banner captures [`billing-banner.txt`](billing-banner.txt) · harness [`floor.sh`](floor.sh)
+
+## Why this exists
+
+An external consumer (the 知音 AI project) benchmarked OCP's prompt path and measured
+**TTFT p50 ≈ 30–32 s**, and excluded OCP as a backend on that basis. That number is real,
+but it is *not* the model being slow — this document decomposes where the 30 seconds
+actually go, and what OCP can do about it.
+
+**The harness deliberately does not go through OCP.** It spawns `tmux` + `claude` directly
+(session prefix `zhiyin-floor-`, never `ocp-tui-*`) and polls `tmux capture-pane` for
+incremental render, so it measures the **true first-token time** of the underlying
+subscription path — the floor OCP could reach if it were perfect.
+
+---
+
+## Measurements
+
+All rows in [`measurements.jsonl`](measurements.jsonl); every number below is recomputable from it.
+
+| Config | n | boot→input-ready (median) | **TTFT (median)** | TTFT range | full answer (median) |
+|---|---|---|---|---|---|
+| baseline (inherits global `effortLevel: xhigh`) | 5 | 1.07 s | **10.35 s** | 8.32 – 17.19 s | 11.32 s |
+| **`--effort low`** | 5 | 1.03 s | **6.17 s** | **5.87 – 6.44 s** | 9.98 s |
+| `--bare` | 5 | 0.44 s | **no answer at all** (5/5 `ttft_ms: -1`) | — | — |
+
+> **Not from this harness**: the direct Anthropic API reference figure (TTFT 0.84–1.64 s, n=2)
+> comes from the 知音 AI project's own smoke test, not from `measurements.jsonl`. It is quoted
+> only to size the gap; do not look for it in the evidence file.
+
+### Where the 30 seconds go
+
+```
+ ~1.0 s   spawn → claude's input bar is ready          ← NOT the bottleneck
+ ~6-10 s  true TTFT (first token rendered in the pane)
+ ~20 s    ████ waiting for the whole turn to finish ████  ← this is the 30s
+```
+
+`runTuiTurn` blocks on the native transcript until a terminal event (`lib/tui/session.mjs`
+"Block on the native transcript … until terminal"; `readTuiTranscript` in
+`lib/tui/transcript.mjs`; ADR 0007 step 4) — i.e. it waits for the **entire turn** to complete
+before returning anything. There is no streaming path. The ~20 s delta between this harness's
+real TTFT and OCP's reported 30–32 s is exactly that.
+
+> **⚠️ 2026-07-13 correction — this decomposition attributes the ~20 s to the wrong thing.** It was
+> inferred from the external 30–32 s report, never measured *through* OCP. It has since been measured
+> through a real OCP instance (TUI mode, `claude-sonnet-4-6`, the same ~1850-token prompt, n=5):
+> **median 11.30 s** before [#156](https://github.com/dtzp555-max/ocp/pull/156), **9.55 s** after.
+> Same-turn decomposition (baseline row `i=5`): **11.563 s** wall through OCP vs `turn_duration:
+> 7.319 s` of CLI-internal time on that same turn → **OCP's own overhead ≈ 4.2 s** (n=1), **not
+> ~20 s**. The rest of any larger number is the model *generating a long answer*,
+> which the blocking wait does not cause and streaming would not shorten — it would only move the
+> first byte earlier. The 30–32 s figure therefore reflects a much longer output (and/or the
+> then-inherited `xhigh` effort), not 20 s of OCP dead time. See
+> [`streaming-spike.md`](streaming-spike.md) § "What streaming would have bought".
+
+---
+
+## ⚠️ Blocking constraint: `--bare` silently drops you off the subscription pool
+
+Captured live ([`billing-banner.txt`](billing-banner.txt)) — the startup banner is the **only**
+reliable indicator:
+
+```
+[]                     | Sonnet 5 with xhigh effort · Claude Max
+[--effort low]         | Sonnet 5 with low effort  · Claude Max
+[--bare]               | Sonnet 5 with xhigh effort · API Usage Billing     ← ❌
+```
+
+`--bare` ("skip hooks, LSP, plugin…") **also skips the subscription-credential resolution
+path**. It really does cut boot to 0.43–0.45 s — but you are no longer on the subscription,
+which defeats the entire purpose of TUI mode (ADR 0007 exists solely to reach the
+subscription pool).
+
+**The failure is silent.** All 5 `--bare` samples reached input-ready (boot 0.43–0.45 s), were
+sent the prompt, and then produced **no answer at all** — 60 s timeout, no error, no crash, the
+pane simply never rendered a token (the API-billing account had no credit balance). Nothing in
+the transcript or the exit status reveals this.
+
+**Anyone changing spawn flags must diff the banner line before and after.**
+
+---
+
+## Backlog — four items, ranked by value ÷ effort
+
+### 1. Pass `--effort` explicitly on spawn — **do this first**
+
+`buildTuiCmd` (`lib/tui/session.mjs`) does not pass `--effort` — `grep -rn -- "--effort\|effortLevel" lib/ server.mjs`
+returns zero hits. What the pane's `claude` ends up using therefore depends on **which HOME mode
+`resolveTuiHome()` picked**:
+
+| mode | HOME | effort the pane gets |
+|---|---|---|
+| **real-home** (legacy default — *current* service config: no `CLAUDE_CODE_OAUTH_TOKEN`, no `OCP_TUI_HOME`) | `~` | **inherits the operator's `~/.claude/settings.json` → `effortLevel: xhigh` on this host** |
+| env-token scratch (`CLAUDE_CODE_OAUTH_TOKEN` set — the direction #146/#150 pushed) | `~/.ocp-tui/home` | that settings.json contains only `permissions.additionalDirectories`; `prepareTuiHome()` never writes `effortLevel` → **claude's built-in default** |
+
+**Scope note**: TUI mode is currently *off* on this host (`CLAUDE_TUI_MODE=false`; `/health` →
+`"tui": {"enabled": false}`), so live traffic takes the `-p` path today. The statement below is
+about what happens **when TUI mode is enabled**.
+
+On the current HOME config, **every TUI request would run extended thinking** — pure waste
+for the typical "generate this JSON" request, and it makes latency depend on an unrelated global
+setting the operator may have changed for their own interactive use. And the mode split means
+the effort level silently changes if the operator ever switches to env-token mode.
+**Passing `--effort` explicitly fixes both problems at once.**
+
+- **Effect (real-home, measured)**: TTFT p50 **10.35 s → 6.17 s (−40 %)**, and the spread
+  collapses from 8.32–17.19 s to **5.87–6.44 s**. For a proxy, the variance reduction matters
+  more than the median.
+- **Cost**: one flag. Suggested: a new `OCP_TUI_EFFORT` env var (default `low`), documented in
+  README § "Environment Variables" per `release_kit.new_feature_doc_expectations`.
+- **Risk**: none — banner confirms it stays on `Claude Max` (see `billing-banner.txt`).
+- ⚠️ Do **not** reach for `--bare` to shave boot: see above.
+
+### 2. Real streaming instead of blocking on turn-terminal — **ACHIEVABLE → [`streaming-spike.md`](streaming-spike.md)**
+
+> **2026-07-13 update — the prereq spike was run. The answer is YES, but not from either source this
+> item guessed at.** (a) The transcript grows at *event* granularity (the whole answer lands in one
+> line, ~0.3 s before terminal) — dead. (b) The pane is a **rendered** view whose `capture-pane` text
+> no longer contains the answer's source bytes (`## `, `**`, code fences are gone) — dead, and worse
+> than "lossy": it is *not the model's text*. **But there is a third source neither this backlog nor
+> the first spike considered: `claude` fires a `MessageDisplay` hook carrying incremental,
+> byte-faithful `delta`s of the raw reply.** Verified live on a plain interactive TUI spawn (no `-p`),
+> banner `· Claude Max`: 7 fires spread across generation, `concat(deltas) === T` **byte-exactly**
+> (579 == 579), `T.startsWith(S)` true at every step, `## ` / `**` / ```` ```javascript ```` all
+> present in the deltas. Granularity is block-level (~5–7 chunks/answer), not token-level — plenty for
+> SSE. **Build it.**
+>
+> ⚠️ Two corrections to this item as written: the **"~20 s" is wrong** (inferred from an external
+> report, never measured through OCP — the same-turn decomposition puts OCP's own overhead at **~4 s**,
+> n=1), and **streaming moves the first byte, not the last** — so a consumer needing the *complete*
+> answer (the JSON-card case that motivated this) gains **nothing** from it. Build it for
+> progressively-rendering consumers, not as a throughput win.
+>
+> Full evidence + implementer caveats (the hook is `forceSyncExecution` — claude BLOCKS on it):
+> **[`streaming-spike.md`](streaming-spike.md)**. Original framing preserved below.
+
+Today `runTuiTurn` blocks on the transcript until the turn is *finished*. The pane is already
+rendering tokens incrementally the whole time — this harness proves you can observe first token
+at ~6 s by polling `tmux capture-pane`.
+
+- **Effect**: turns a 30 s wall into a ~6 s TTFT with progressive output; enables SSE streaming
+  on the OCP endpoint instead of a single blob at the end.
+- **Cost**: real work. Pane capture is ANSI/redraw-based and lossy for exact text (wrapping,
+  scrollback, spinner lines). Two candidate sources: (a) incremental reads of the transcript
+  JSONL, (b) `capture-pane` diffing with a stable start marker. (a) is much cleaner **if it
+  holds**.
+- **Prereq spike (do this before designing anything)**: does the transcript JSONL grow *during*
+  a turn, or only at the end? If only at the end, (a) is dead and you are stuck with (b).
+
+### 3. Warm pane pool — ~1 s
+
+Every request spawns a fresh tmux session + `claude` (`randomUUID()` + `new-session`, then
+`kill-session` in `finally`; `grep -rn "pool\|warm\|reuse" lib/tui/*.mjs` → zero hits). Boot to
+input-ready is ~1.0 s, paid on every request. A pool of pre-booted panes (single-use, replaced in
+the background) amortizes it to zero for any workload below the pool refill rate.
+
+- **Effect**: −1.0 s.
+- **Cost**: moderate; interacts with the session reaper and the per-port prefix scoping added in
+  #148 — pooled panes must not look like zombies to the sweep.
+- Lower priority than #1 and #2: it is the smallest slice.
+
+### 4. Trim the prefill — ~~probably not worth it~~ **MEASURED: no detectable benefit. Do not adopt.**
+
+> **2026-07-13 update.** `--exclude-dynamic-system-prompt-sections` was measured with the same
+> harness (`floor.sh`, n=5, Sonnet 5, on top of `--effort low`): **TTFT median 6.39 s**
+> (5.87–10.54 s) vs **6.17 s** (5.87–6.44 s) for `--effort low` alone — i.e. **0.22 s worse, inside
+> the noise band**, with one worse outlier; dropping that outlier does not change the verdict. n=5
+> cannot prove "zero", only "no benefit detectable above noise" — but there is also a **mechanistic**
+> reason not to expect one: `--help` says the flag *"Improves cross-user prompt-cache **reuse**"*, and
+> **OCP is single-user** — there is no cross-user cache to share, so the flag has nothing to buy here.
+> The banner stayed on `· Claude Max` (no billing-pool drop), but there is no win to bank. The ~6 s
+> floor stands as stated below. Raw rows: [`prefill-spike-measurements.jsonl`](prefill-spike-measurements.jsonl).
+
+
+After #1–#3, the floor is **~6 s**, and it does not go lower. `claude` always injects the full
+Claude Code system prompt + tool definitions (thousands to tens of thousands of prefill tokens)
+regardless of what you ask it. `--exclude-dynamic-system-prompt-sections` exists and may shave
+some of it — **unmeasured**; worth one spike, but do not expect to reach the direct API's
+~1 s.
+
+**Consequence to accept, and to state in the README**: even fully optimized, TUI mode has a
+**~6 s TTFT floor**, so it cannot serve real-time / interactive-latency consumers. It remains
+appropriate for batch, background, and cost-insensitive-latency use. The 知音 AI project
+excluded it on this basis (their prompt-latency budget is 2–4 s) *independently* of the ToS
+question already documented in the README.
+
+---
+
+## Reproduction
+
+```bash
+# harness never touches OCP's :3456 service or ocp-tui-* sessions, and never kill-server
+bash docs/plans/2026-07-13-tui-latency/floor.sh 5                                  # baseline
+TAG=effort-low EXTRA_ARGS="--effort low" bash .../floor.sh 5                       # −40 %
+TAG=bare       EXTRA_ARGS="--bare"       bash .../floor.sh 5                       # the trap
+
+# billing-pool check for ANY spawn-flag change — the banner is the only source of truth
+tmux new-session -d -s probe -x 200 -y 50 -c "$HOME" \
+  "claude --model claude-sonnet-5 --session-id $(uuidgen) <your-flags-here>"
+sleep 6; tmux capture-pane -p -t probe | grep -E "Claude Max|API Usage Billing"
+tmux kill-session -t probe
+```
+
+## Interaction with OCP while the harness runs
+
+- **Kill direction is safe both ways**: `reapStaleTuiSessions()` only `kill-session`s names
+  matching `ocp-tui-<port>-`, which `zhiyin-floor-*` never matches; and the harness only
+  `kill-session`s its own single session — it contains **no `kill-server`**.
+- **One benign interaction** (only when TUI mode is enabled — the reap tick is itself gated on
+  `TUI_MODE`): OCP's periodic `kill-server` (zombie reaping) is gated on
+  `othersRemain` — *any* foreign-prefixed tmux session suppresses it. So while the harness is
+  running, that sweep is skipped. This is the coexistence guard working as designed; it resumes
+  on the next tick.
+
+## Harness caveats (stated so the numbers are not over-trusted)
+
+- **n=5 per config**, single host, single model (Sonnet 5), single prompt size (~1850 tokens).
+  Enough to separate 6 s from 10 s from 30 s; **not** enough for a p95.
+- TTFT is "marker visible in `capture-pane`", which includes tmux render latency (small, but
+  nonzero) — it is an upper bound on the true first-token time.
+- **The harness's readiness marker is not OCP's.** `floor.sh` waits for `│ >|❯|Try "`; OCP's
+  `tuiInputReady()` matches `/\? for shortcuts/`. These are different events, so the ~1.0 s
+  boot figure is **not** directly comparable to OCP's `BOOT_MS` gate (default cap 4000 ms). It
+  does not affect the conclusions (1 s ≪ 6 s TTFT), but it is not apples-to-apples.
+- The first version of this harness reported TTFT **0.08 s** — a false positive: the prompt
+  literally contained the marker string it was grepping for, so the match fired the instant the
+  prompt was pasted. Fixed by describing the marker instead of spelling it. **The script exited 0
+  and "successfully" produced 5 samples both times** — exit status proves nothing here.
@@ -0,0 +1,3 @@
+[]                     | ▝▜█████▛▘  Sonnet 5 with xhigh effort · Claude Max
+[--effort low]         | ▝▜█████▛▘  Sonnet 5 with low effort · Claude Max
+[--bare]               | ▝▜█████▛▘  Sonnet 5 with xhigh effort · API Usage Billing
@@ -0,0 +1,128 @@
+#!/usr/bin/env bash
+# OCP TUI-mode latency floor harness — see README.md in this directory.
+#
+# 目的：回答一个问题——如果把 OCP 现有的两个已知开销砍掉
+#   (a) 每请求 spawn + boot（可用预热进程池消除）
+#   (b) 假流式（等 turn_duration 才返回，可用增量读 pane 消除）
+# 之后，订阅池路径的**真实 TTFT 地板**是多少？
+#
+# 判据：地板 ≤ 4s → OCP 作为"省钱选项"可行；> 8s → 死透，不再讨论。
+#
+# 红线：
+# - 不经过生产 OCP 服务（:3456）—— 直接起 tmux+claude，OCP 进程零干扰
+# - tmux session 前缀用 zhiyin-floor-（**不是** ocp-tui-），避免被 OCP 的
+#   reaper 当成自己的会话杀掉，也避免我们杀到它的
+# - 用 real HOME（凭据）—— scratch HOME + symlink 凭据会 fork OAuth 导致 401
+#   （见跨机记忆 tui_scratch_home_credential_fork）
+set -uo pipefail
+
+N=${1:-5}
+MODEL=${MODEL:-claude-sonnet-5}
+EXTRA_ARGS=${EXTRA_ARGS:-}          # 额外 CLI 参数（如 --effort low --bare）
+TAG=${TAG:-baseline}
+OUT=${OUT:-$(dirname "$0")/measurements.jsonl}
+PROMPT_FILE=$(mktemp)
+PREFIX="zhiyin-floor"
+
+mkdir -p "$(dirname "$OUT")"
+
+# ── 构造提示：~2000 token 的假会议转写 + 明确的起始标记 ────────────────
+# 单行（多行会在 tmux send-keys 时提前触发 Enter）
+build_prompt() {
+  local seg="Speaker A said the quarterly pipeline is tracking behind plan and the enterprise segment needs a different motion. Speaker B replied that the current onboarding flow loses roughly a third of trial accounts before the first integration is complete. They debated whether the fix belongs in product or in customer success. "
+  local body=""
+  for _ in $(seq 1 22); do body+="$seg"; done
+  printf '%s' "You are a real-time meeting copilot. Meeting transcript so far: $body --- Task: produce ONE prompt card as compact JSON with keys: points (array of 3 short Chinese bullet points), keyline (one English sentence the user can read aloud). IMPORTANT: your reply MUST begin with three hash characters immediately followed by the uppercase word CARD (no space between them), then the JSON. No preamble, no markdown fences." > "$PROMPT_FILE"
+}
+build_prompt
+PROMPT_CHARS=$(wc -c < "$PROMPT_FILE" | tr -d ' ')
+
+now_ms() { python3 -c 'import time;print(int(time.time()*1000))'; }
+
+echo "配置: $TAG   参数: [$EXTRA_ARGS]"
+echo "模型: $MODEL   样本: $N   提示长度: ${PROMPT_CHARS} chars (≈$((PROMPT_CHARS/4)) token)"
+echo "输出: $OUT"
+echo
+
+for i in $(seq 1 "$N"); do
+  SESS="${PREFIX}-$$-$i"
+  SID=$(uuidgen)
+
+  # ── 冷启动：spawn + 等输入框就绪 ─────────────────────────────────
+  T_SPAWN=$(now_ms)
+  tmux new-session -d -s "$SESS" -x 200 -y 50 \
+    -e CLAUDE_CODE_DISABLE_CLAUDE_MDS=1 \
+    -e CLAUDE_CODE_DISABLE_AUTO_MEMORY=1 \
+    -c "$HOME" \
+    "claude --model $MODEL --session-id $SID --strict-mcp-config --disallowedTools 'mcp__*' $EXTRA_ARGS" 2>/dev/null
+  if [ $? -ne 0 ]; then echo "[$i] tmux spawn 失败，跳过"; continue; fi
+
+  # 轮询输入框就绪（claude TUI 的输入提示符）
+  READY=0
+  for _ in $(seq 1 150); do   # 上限 15s
+    PANE=$(tmux capture-pane -p -t "$SESS" 2>/dev/null || true)
+    if grep -qE '│ >|❯|Try "' <<<"$PANE"; then READY=1; break; fi
+    sleep 0.1
+  done
+  T_READY=$(now_ms)
+  BOOT_MS=$((T_READY - T_SPAWN))
+  if [ "$READY" -ne 1 ]; then
+    echo "[$i] 启动超时（${BOOT_MS}ms），pane 末 3 行:"
+    tmux capture-pane -p -t "$SESS" 2>/dev/null | tail -3 | sed 's/^/      /'
+    tmux kill-session -t "$SESS" 2>/dev/null
+    continue
+  fi
+
+  # ── 热态：粘提示 → 回车 → 量首 token ─────────────────────────────
+  tmux send-keys -t "$SESS" -l "$(cat "$PROMPT_FILE")" 2>/dev/null
+  sleep 0.4                       # 让粘贴落地（OCP 用 400ms 轮询粒度）
+  T0=$(now_ms)
+  tmux send-keys -t "$SESS" Enter 2>/dev/null
+
+  TTFT_MS=-1
+  for _ in $(seq 1 600); do       # 上限 60s
+    if tmux capture-pane -p -t "$SESS" 2>/dev/null | grep -q '###CARD'; then
+      TTFT_MS=$(( $(now_ms) - T0 )); break
+    fi
+    sleep 0.1
+  done
+
+  # ── 完整回答：pane 连续 2s 不再变化 ──────────────────────────────
+  COMPLETE_MS=-1
+  if [ "$TTFT_MS" -ge 0 ]; then
+    LAST=""; STABLE=0
+    for _ in $(seq 1 900); do     # 上限 90s
+      CUR=$(tmux capture-pane -p -t "$SESS" 2>/dev/null | cksum)
+      if [ "$CUR" = "$LAST" ]; then
+        STABLE=$((STABLE+1))
+        [ "$STABLE" -ge 20 ] && { COMPLETE_MS=$(( $(now_ms) - T0 - 2000 )); break; }
+      else
+        STABLE=0; LAST="$CUR"
+      fi
+      sleep 0.1
+    done
+  fi
+
+  printf '{"i":%d,"tag":"%s","model":"%s","extra_args":"%s","prompt_chars":%s,"boot_ms":%d,"ttft_ms":%d,"complete_ms":%d}\n' \
+    "$i" "$TAG" "$MODEL" "$EXTRA_ARGS" "$PROMPT_CHARS" "$BOOT_MS" "$TTFT_MS" "$COMPLETE_MS" | tee -a "$OUT"
+
+  tmux kill-session -t "$SESS" 2>/dev/null
+  sleep 1
+done
+
+rm -f "$PROMPT_FILE"
+echo
+echo "=== 汇总 ==="
+python3 - "$OUT" <<'EOF'
+import json,sys,statistics
+rows=[json.loads(l) for l in open(sys.argv[1]) if l.strip()]
+ok=[r for r in rows if r['ttft_ms']>=0]
+if not ok: print("无有效样本"); sys.exit()
+def s(k):
+    v=[r[k] for r in ok if r[k]>=0]
+    return f"n={len(v)} 中位={statistics.median(v)/1000:.2f}s 最小={min(v)/1000:.2f}s 最大={max(v)/1000:.2f}s" if v else "无"
+print(f"  冷启动 boot      : {s('boot_ms')}   ← 预热进程池可完全消除")
+print(f"  TTFT（首 token） : {s('ttft_ms')}   ★ 这就是地板")
+print(f"  完整回答         : {s('complete_ms')}")
+print(f"\n  失败样本: {len(rows)-len(ok)}/{len(rows)}")
+EOF
@@ -0,0 +1,15 @@
+{"i": 1, "tag": "effort-low", "model": "claude-sonnet-5", "extra_args": "--effort low", "prompt_chars": 7451, "boot_ms": 1077, "ttft_ms": 6172, "complete_ms": 9929}
+{"i": 2, "tag": "effort-low", "model": "claude-sonnet-5", "extra_args": "--effort low", "prompt_chars": 7451, "boot_ms": 1026, "ttft_ms": 6160, "complete_ms": 9996}
+{"i": 3, "tag": "effort-low", "model": "claude-sonnet-5", "extra_args": "--effort low", "prompt_chars": 7451, "boot_ms": 1010, "ttft_ms": 6437, "complete_ms": 9977}
+{"i": 4, "tag": "effort-low", "model": "claude-sonnet-5", "extra_args": "--effort low", "prompt_chars": 7451, "boot_ms": 1033, "ttft_ms": 5872, "complete_ms": 9944}
+{"i": 5, "tag": "effort-low", "model": "claude-sonnet-5", "extra_args": "--effort low", "prompt_chars": 7451, "boot_ms": 1154, "ttft_ms": 6387, "complete_ms": 9993}
+{"i":1,"tag":"baseline","model":"claude-sonnet-5","extra_args":"","prompt_chars":7451,"boot_ms":1300,"ttft_ms":8321,"complete_ms":9939}
+{"i":2,"tag":"baseline","model":"claude-sonnet-5","extra_args":"","prompt_chars":7451,"boot_ms":1070,"ttft_ms":10347,"complete_ms":11320}
+{"i":3,"tag":"baseline","model":"claude-sonnet-5","extra_args":"","prompt_chars":7451,"boot_ms":911,"ttft_ms":13061,"complete_ms":15163}
+{"i":4,"tag":"baseline","model":"claude-sonnet-5","extra_args":"","prompt_chars":7451,"boot_ms":1441,"ttft_ms":9981,"complete_ms":11066}
+{"i":5,"tag":"baseline","model":"claude-sonnet-5","extra_args":"","prompt_chars":7451,"boot_ms":1036,"ttft_ms":17189,"complete_ms":17985}
+{"i":1,"tag":"bare","model":"claude-sonnet-5","extra_args":"--bare","prompt_chars":7451,"boot_ms":429,"ttft_ms":-1,"complete_ms":-1}
+{"i":2,"tag":"bare","model":"claude-sonnet-5","extra_args":"--bare","prompt_chars":7451,"boot_ms":437,"ttft_ms":-1,"complete_ms":-1}
+{"i":3,"tag":"bare","model":"claude-sonnet-5","extra_args":"--bare","prompt_chars":7451,"boot_ms":444,"ttft_ms":-1,"complete_ms":-1}
+{"i":4,"tag":"bare","model":"claude-sonnet-5","extra_args":"--bare","prompt_chars":7451,"boot_ms":446,"ttft_ms":-1,"complete_ms":-1}
+{"i":5,"tag":"bare","model":"claude-sonnet-5","extra_args":"--bare","prompt_chars":7451,"boot_ms":441,"ttft_ms":-1,"complete_ms":-1}
@@ -0,0 +1,7 @@
+{"hook_event_name": "MessageDisplay", "index": 0, "final": false, "delta": "## Mutex\n\n"}
+{"hook_event_name": "MessageDisplay", "index": 1, "final": false, "delta": "A **mutual exclusion lock** prevents concurrent access to a shared resource, ensuring only one thread runs the critical section at a time.\n\n"}
+{"hook_event_name": "MessageDisplay", "index": 2, "final": false, "delta": "- Acquiring a locked mutex blocks the caller until the current holder releases it.\n"}
+{"hook_event_name": "MessageDisplay", "index": 3, "final": false, "delta": "- Failing to release a mutex causes a deadlock, freezing all waiting threads.\n\n```javascript\nconst { Mutex } = require('async-mutex');\n\nconst mutex = new Mutex();\n"}
+{"hook_event_name": "MessageDisplay", "index": 4, "final": false, "delta": "let counter = 0;\n\nasync function increment() {\n  const release = await mutex.acquire();\n  try {\n"}
+{"hook_event_name": "MessageDisplay", "index": 5, "final": false, "delta": "    counter++; // only one caller here at a time\n  } finally {\n    release();\n  }\n}\n"}
+{"hook_event_name": "MessageDisplay", "index": 6, "final": true, "delta": "```"}
@@ -0,0 +1,5 @@
+{"i":1,"tag":"effort-low-exclude-dynamic","model":"claude-sonnet-5","extra_args":"--effort low --exclude-dynamic-system-prompt-sections","prompt_chars":7451,"boot_ms":934,"ttft_ms":5867,"complete_ms":9953}
+{"i":2,"tag":"effort-low-exclude-dynamic","model":"claude-sonnet-5","extra_args":"--effort low --exclude-dynamic-system-prompt-sections","prompt_chars":7451,"boot_ms":1275,"ttft_ms":6388,"complete_ms":9874}
+{"i":3,"tag":"effort-low-exclude-dynamic","model":"claude-sonnet-5","extra_args":"--effort low --exclude-dynamic-system-prompt-sections","prompt_chars":7451,"boot_ms":874,"ttft_ms":10537,"complete_ms":11782}
+{"i":4,"tag":"effort-low-exclude-dynamic","model":"claude-sonnet-5","extra_args":"--effort low --exclude-dynamic-system-prompt-sections","prompt_chars":7451,"boot_ms":1170,"ttft_ms":6379,"complete_ms":9947}
+{"i":5,"tag":"effort-low-exclude-dynamic","model":"claude-sonnet-5","extra_args":"--effort low --exclude-dynamic-system-prompt-sections","prompt_chars":7451,"boot_ms":1329,"ttft_ms":6443,"complete_ms":9884}
@@ -0,0 +1,256 @@
+# Backlog #2 (real streaming): **achievable** — via the `MessageDisplay` hook
+
+**Date**: 2026-07-13
+**Status**: prereq-spike result. **Streaming IS achievable on the TUI path**, byte-faithfully, on the
+subscription pool. Three obvious sources are dead ends; a fourth one works.
+**Scope**: answers the prereq spike that [`README.md`](README.md) § "Backlog #2" demanded *before* any
+streaming design:
+
+> **Prereq spike (do this before designing anything)**: does the transcript JSONL grow *during* a
+> turn, or only at the end? If only at the end, (a) is dead and you are stuck with (b).
+
+The answer: **(a) is dead, (b) is dead — and you are not stuck with either.** The CLI exposes its own
+streaming interface as a **hook**, which the backlog did not consider.
+
+**Measured on**: Mac mini / Claude Code **v2.1.207** / Sonnet 4.6 + Sonnet 5 / Claude Max /
+real-home mode. Every claim below is reproducible from the commands given.
+
+> **Honesty note on how this document was produced.** Its first version concluded the exact opposite —
+> "streaming is not achievable; the CLI exposes no byte-faithful incremental source" — and was **wrong**.
+> An adversarial reviewer, commissioned specifically to *refute* it, found `MessageDisplay` on a second
+> pass; its own first pass had enumerated the hook registry with a truncated grep (it reported 21
+> events — there are **30**). Both the wrong conclusion and its refutation are preserved here, because
+> "we checked, it's impossible" is the most expensive kind of claim to get wrong: it closes a door and
+> nobody re-opens it.
+
+---
+
+## ✅ The source that works: the `MessageDisplay` hook
+
+`claude` fires a **`MessageDisplay`** hook as it renders each block of the assistant's reply. The
+payload carries the **raw markdown source** of an incremental `delta`, plus a monotonic `index` and a
+`final` flag:
+
+```json
+{ "hook_event_name": "MessageDisplay",
+  "turn_id": "6cb31d21-…", "message_id": "84ab9832-…",
+  "index": 0, "final": false, "delta": "## Mutex\n\n" }
+```
+*(payload also carries `session_id`, `transcript_path`, `prompt_id`, `cwd`)*
+
+Registered as an ordinary command hook via `--settings` on a **plain interactive TUI spawn** (no `-p`,
+no `--bare`), `claude-sonnet-4-6`, `--effort low`. Banner verified:
+`▝▜█████▛▘ Sonnet 4.6 with low effort · Claude Max` — **subscription pool, not metered billing**.
+
+One live turn — 7 fires, spread across generation:
+
+```
+index=0 final=false  len=  10  '## Mutex\n\n'
+index=1 final=false  len= 140  'A **mutual exclusion lock** prevents concurrent access to a shar…'
+index=2 final=false  len=  83  '- Acquiring a locked mutex blocks the caller until the current h…'
+index=3 final=false  len= 163  '- Failing to release a mutex causes a deadlock, freezing all wai…'
+index=4 final=false  len=  96  'let counter = 0;\n\nasync function increment() {\n  const release =…'
+index=5 final=false  len=  84  '    counter++; // only one caller here at a time\n  } finally {\n …'
+index=6 final=true   len=   3  '```'
+```
+
+**Every invariant a proxy needs — all hold:**
+
+| requirement | result |
+|---|---|
+| **byte-faithful** — deltas are the model's *source*, not the rendered pane | ✅ `## `, `**`, ```` ```javascript ```` all present in the deltas |
+| **exactness** — `concat(deltas) === T` (the transcript-authoritative text) | ✅ **true**, 579 == 579 bytes |
+| **prefix-stable** — `T.startsWith(concat(deltas[0..n]))` at every n | ✅ **true at all 7 steps** |
+| **incremental** — arrives during generation, not at the end | ✅ 7 fires spread across the turn |
+| **no `-p`** — stays out of the metered `sdk-cli` pool | ✅ plain interactive TUI |
+| **subscription pool** | ✅ banner `· Claude Max` |
+
+This is exactly the contract a streaming design needs: deltas forward straight into SSE
+`delta.content` chunks, and the transcript's final text `T` stays a cheap end-of-turn assertion
+(`concat === T`) instead of a reconciliation problem.
+
+### Caveats for the implementer
+
+- **Block-level granularity, not token-level** — the hook fires **once per rendered block** (roughly one
+  per paragraph / list item / code block), so the chunk count **scales with answer length**: 7 fires for a
+  ~600-byte answer, **18 for a ~2 KB one**. Plenty for SSE (`delta.content` has no minimum size), but do
+  not promise token-by-token output, and do not hard-code any assumption about chunk count.
+- **🔴 The sink MUST be keyed by `session_id` — this is live TODAY, not a future concern.**
+  `OCP_TUI_MAX_CONCURRENT` defaults to **2**, so **two `claude` processes already run concurrently**. One
+  hook command writing to one shared sink would **interleave deltas from two different turns into one
+  stream** — request A's client receiving request B's text, the worst failure a proxy can have, and one a
+  single-request test will never surface. The payload carries `session_id` (and `turn_id` / `message_id`),
+  so demux is easy: derive the sink path from `session_id` (`<dir>/<session_id>.jsonl`) and read only your
+  own turn's file. This *also* keeps the design **warm-pool compatible**, because a pre-booted pane's
+  session-id is fixed at boot — one static hook script serves every pane. **Test it with ≥2 concurrent
+  streaming requests carrying distinguishable prompts and assert zero cross-contamination.**
+- **⚠️ `forceSyncExecution: true` in the hook's source — `claude` BLOCKS on the hook.** A slow hook
+  adds latency to *every* delta. The hook must write and exit immediately (e.g. write to a FIFO / unix
+  socket that OCP reads; never work inline). **Measure the added per-delta latency.**
+- **Thinking blocks appear to be excluded — but this is NOT yet stress-tested. Verify before shipping.**
+  The exclusion is inferred from `content.map(c => c.type === "text" ? c.text : "")` — but that snippet is
+  from the **`final:true`** call site, not the incremental one. Four live turns (incl. two at `--effort
+  high`) showed no thinking text in any delta and `concat === T` held — **but each transcript's thinking
+  block was empty (`thinking:""`, 0 chars)**, so the exclusion was never actually stressed. **The failure
+  mode is severe**: if thinking deltas *do* fire on some config (Opus, `xhigh`), `concat(deltas) !== T`
+  **and OCP streams the model's private reasoning to the caller**. The end-of-turn `concat === T` assertion
+  would *detect* that but **cannot prevent** it — SSE deltas cannot be un-sent. **Before shipping, run a
+  turn on a model+effort that produces substantive thinking** (a hard reasoning prompt on Opus / `xhigh`)
+  and confirm both (a) no thinking text in any delta and (b) `concat === T` still holds.
+- OCP already owns the spawn (isolated HOME, its own flags), so injecting `--settings` with a
+  `MessageDisplay` hook sits inside the existing architecture.
+- **`ALIGNMENT.md`**: this consumes `claude`'s **own** hook surface as emitted — forwarding, not
+  inventing. Not a new endpoint, not a fabricated protocol. (Class B / ADR 0007 — the TUI spawn is
+  OCP-owned; no `cli.js` citation applies.)
+
+### Reproduce in 60 seconds
+
+```bash
+# hook script: append the payload (arrives on stdin) and exit immediately
+printf '#!/bin/bash\ncat >> "$MD_LOG"; printf "\\n" >> "$MD_LOG"; exit 0\n' > /tmp/h.sh && chmod +x /tmp/h.sh
+echo '{"hooks":{"MessageDisplay":[{"hooks":[{"type":"command","command":"MD_LOG=/tmp/deltas.jsonl /tmp/h.sh"}]}]}}' > /tmp/s.json
+
+# plain interactive claude in tmux (prefix NOT ocp-tui-*, and never kill-server)
+tmux new-session -d -s md-probe -x 220 -y 50 \
+  "claude --model claude-sonnet-4-6 --effort low --session-id $(uuidgen) --settings /tmp/s.json"
+# …wait for '? for shortcuts', paste a markdown-producing prompt, press Enter…
+
+jq -r '"\(.index) \(.final) \(.delta|@json)"' /tmp/deltas.jsonl   # incremental raw-markdown deltas
+# then assert: concat(deltas) == extractLatestAssistantText(<transcript>.jsonl)
+```
+
+---
+
+## The three dead ends (still worth knowing — they say what NOT to build)
+
+### (a) Incremental transcript reads — **dead: event granularity, not token granularity**
+
+The transcript JSONL *does* grow during a turn, but one **whole event at a time**; the assistant's text
+event is written as **one complete line**, appearing only ~0.3 s before the terminal `turn_duration`.
+
+Observed (session `efd5b161`, `turn_duration: 7319 ms`):
+
+```
+ #6  t+0.0s   type=user       (the prompt)
+ #15 t+4.7s   type=assistant  blocks=thinking
+ #16 t+7.0s   type=assistant  blocks=text      ← the ENTIRE answer, in one line
+ #21 t+7.3s   type=system     subtype=turn_duration   ← terminal
+```
+
+Cross-checked at **20 ms polling + `fs.watch`** (25× finer): a partial line **never touches disk** —
+one write, `+1` line, carrying the complete answer. Also forced with the undocumented
+`CLAUDE_CODE_INCLUDE_PARTIAL_MESSAGES=1`: still 1 assistant event, 0 partials (interactive mode has no
+stream-json *sink* for it to write to).
+
+**The transcript is still needed** — as the terminal-turn signal, as the authoritative `concat === T`
+check, and as the input to the existing honesty gates (auth-banner detection, `truncated`). It is just
+not the *streaming* source.
+
+### (b) `tmux capture-pane` diffing — **dead: the pane is a RENDERED view, not the text**
+
+The backlog expected to fall back to this, calling it "lossy … (wrapping, scrollback, spinner lines)".
+The loss is far worse than formatting noise: **the pane does not contain the answer's source bytes at
+all.** The TUI *renders* markdown, and `capture-pane -p` strips the ANSI that rendering produced.
+
+Same turn, same lines:
+
+```
+TRANSCRIPT (authoritative T):          PANE (capture-pane -p -J -S -500):
+  '## Semaphore'                         '⏺ Semaphore'                 ← heading marker gone
+  ''                                     ''
+  'A **semaphore** is a synchro…'        '  A semaphore is a synchro…' ← bold markers gone, indented
+```
+
+| token in the answer | in `T` | in the pane's answer region |
+|---|---|---|
+| `## ` (ATX heading) | yes | **no** — rendered as `⏺` |
+| `**` (bold markers) | yes | **no** — rendered to ANSI bold, then stripped by `-p` |
+| ` ```javascript ` (fence + language) | yes | **no** — fence and language tag both gone |
+| `- ` (list item) | yes | yes |
+
+*(A literal `**` does appear elsewhere in the pane — in the **prompt echo**, because the prompt asked
+for bold. Not in the answer.)*
+
+**`capture-pane -e` (keeping the ANSI) does not rescue it — the inverse is provably non-unique.**
+With `T` = ``"## Alpha\n\n**bravo**\n\n```javascript\nlet x=1;\n```"``:
+
+```
+⏺\e[39m \e[1mAlpha\n\n\e[0m  \e[1mbravo\n\n\e[0m  \e[34mlet\e[39m x=\e[32m1\e[39m;
+```
+
+`## Alpha` → **SGR 1 (bold)**. `**bravo**` → **SGR 1 (bold)**. *Identical ANSI* — an H2 and a bold span
+are indistinguishable, never mind `**` vs `__`. The fence and its `javascript` tag are consumed by the
+syntax highlighter into colours; recovering the tag would mean inverting a highlighter, and
+`let x=1;` is valid in several languages.
+
+So `T.startsWith(paneText)` is **false** — raw and indent-stripped, on essentially every markdown
+answer. A proxy streaming pane text would be streaming **something the model did not say**. With
+`MessageDisplay` available there is no reason to go near it.
+
+### (c) `--debug-file` — **dead: it logs stream *timing*, never stream *content***
+
+Worth stating precisely, because a casual check misleads in **both** directions here.
+
+The default log level is `debug`, which **suppresses every `verbose` site**. Raise it and per-chunk
+lines *do* appear, spread across generation:
+
+```bash
+CLAUDE_CODE_DEBUG_LOG_LEVEL=verbose claude --debug-file /tmp/d.log …
+```
+```
+05:51:11.088 [VERBOSE] [shoji-engine] yield stream_event/-     ← 16 of these, mid-turn,
+05:51:11.537 [VERBOSE] [shoji-engine] yield stream_event/-        over ~3.9 s of generation
+05:51:15.192 [DEBUG]   [shoji-engine] turn 1 end (usage in=575 out=255 api=6736ms stop=end_turn resultLen=857)
+```
+
+**But they carry no payload** — the format is `yield <type>/<subtype>`, a bare presence marker. Run with
+no category filter (i.e. all categories) at verbose level: `content_block_delta` = **0**, `text_delta` =
+**0**, `content_block_start` / `message_start` = **0**. The only byte-exact text in the log is the
+end-of-turn `Stop` hook payload (`"last_assistant_message":"## Title\n\n**alpha bravo charlie**"`) —
+transcript granularity. The log tells you **when** tokens arrive, never **what** they are. It is also
+~2.7 MB per turn.
+
+### Also checked, also not the answer
+
+| candidate | outcome |
+|---|---|
+| `--output-format stream-json` (the one interface that emits `text_delta`) | **requires `--print`/`-p`** → `cc_entrypoint=sdk-cli` → the **metered** credit pool, which is exactly what TUI mode exists to avoid. Reproduced live. |
+| `--input-format stream-json` | `Error: --input-format=stream-json requires output-format=stream-json` → same gate. |
+| `CLAUDE_CODE_INCLUDE_PARTIAL_MESSAGES=1` (undocumented) | No stream-json sink in interactive mode → no partials. Banner stayed `· Claude Max`. |
+| `sessionMirror` (undocumented) | Gated on `outputFormat === "stream-json"` → the `-p` family. |
+| `--sdk-url` (hidden) | Forces stream-json + non-interactive → `sdk-cli`. *(inferred from the minified bundle; not banner-tested)* |
+| `~/.claude/sessions/<pid>.json` | Registry metadata only (`{pid, sessionId, cwd, status, version, entrypoint:"cli", kind:"interactive"}`). No assistant text. *(Its `entrypoint:"cli"` incidentally confirms the TUI path stays on the subscription pool.)* |
+| `~/.claude/history.jsonl` | User prompts only; the answer text is absent. |
+| Asking the model to emit plain text (so the pane renders faithfully) | Would mean **mutating the caller's prompt** — a correctness violation for a proxy, and still not byte-faithful (wrapping + indent remain). Rejected. |
+
+---
+
+## Value: what streaming actually buys (read before building)
+
+Streaming is *possible*. Whether it is *worth it* depends on the consumer, and the honest answer is
+uncomfortable:
+
+- **Streaming never makes the answer arrive sooner. It moves the *first* byte, not the *last*.** The
+  final token lands at the same wall-clock moment either way.
+- So a consumer that must have the **complete** answer before it can act — e.g. one parsing a structured
+  JSON reply, **which is exactly the 知音 AI use case that motivated this entire investigation** — gains
+  **nothing at all**. Only a **progressively-rendering** consumer (a chat UI) gains.
+
+And the number the backlog attached to this item was wrong:
+
+- The backlog's "~20 s" was inferred from an external 30–32 s report, **never measured through OCP**.
+  Measured through a real OCP instance (TUI mode, `claude-sonnet-4-6`, ~1850-token prompt, n=5):
+  **median 11.30 s** before [#156](https://github.com/dtzp555-max/ocp/pull/156), **9.55 s** after.
+- **Same-turn decomposition** (baseline row `i=5`): **11.563 s** wall through OCP vs `turn_duration:
+  7.319 s` of CLI-internal time on that same turn → **OCP's own overhead ≈ 4.2 s** (n=1, baseline
+  `effort=high` config). *Caveats*: n=1; and `turn_duration` is the CLI's internal duration of an
+  **OCP-driven** turn, not a separate "native" baseline. Do **not** subtract this `effort=high` 7.3 s
+  from the `effort=low` 9.55 s median — a low-effort turn generates faster, so mixing them
+  *understates* the overhead.
+- So OCP's own overhead is **single-digit seconds**, not ~20 s. The rest of any large number is the
+  model generating a long answer — which streaming hides but does not shorten.
+
+**Recommendation**: build it — the contract is clean and the cost is small — but size the expectation
+honestly. It is a *perceived-latency* feature for progressively-rendering consumers, not a throughput
+win, and it does not move the **~6 s TTFT floor** ([`README.md`](README.md)) that rules TUI mode out for
+interactive-latency consumers regardless.
@@ -0,0 +1,321 @@
+import { rmSync } from "node:fs";
+
+// TUI warm pane pool (docs/plans/2026-07-13-tui-latency backlog #3).
+//
+// WHAT IT IS: a small set of PRE-BOOTED `claude` panes, each already sitting at its
+// input bar, so a request does not pay the cold boot. Opt-in: OCP_TUI_POOL_SIZE=0
+// (default) disables it entirely and the request path is byte-for-byte today's.
+//
+// ── SINGLE-USE IS THE LOAD-BEARING RULE ─────────────────────────────────────
+// A pooled pane serves EXACTLY ONE turn and is then killed and replaced in the
+// background. Each pane carries its OWN fresh `--session-id`, fixed at boot, and the
+// turn locates its transcript by that id. So OCP's one-session-per-request model is
+// preserved: a session's transcript still holds exactly one logical exchange.
+// That is what keeps lib/tui/transcript.mjs's extractLatestAssistantText (which returns
+// the LAST text-bearing assistant entry in the whole file, not "text since the matching
+// user line") correct — see the scoping note there. A pane MUST NEVER serve a second
+// turn, and a session MUST NEVER be reset with /clear and reused: either would put two
+// exchanges in one transcript and leak the earlier turn's text into the later turn's
+// answer. Nothing here reuses a pane; keep it that way.
+//
+// ── WHY IT'S WORTH MORE THAN THE BOOT TIME ──────────────────────────────────
+// Measured on this host (n=6 through OCP, Sonnet 4.6, --effort low): the cold path
+// spends ~1.23 s reaching the input bar, but ALSO ~2.9 s inside the first turn beyond
+// what claude itself reports as the turn duration — post-input-bar init that a pane
+// which has been idle for a few seconds has already finished. A warm pane recovers both.
+//
+// ── COST (bounded, and paid whether or not a request arrives) ───────────────
+// Each warm pane is a LIVE `claude` process (plus its tmux pane) sitting idle. Peak
+// process count is (pool size) + (OCP_TUI_MAX_CONCURRENT in-flight turns) + (panes
+// currently booting as replacements). Pool size is clamped to POOL_MAX_SIZE.
+//
+// Pure + injectable (bootPane / killPane / paneHealthy / now) so test-features.mjs can
+// assert acquire / miss / refill / TTL / reaper-exemption with no tmux and no claude.
+
+// Hard cap on OCP_TUI_POOL_SIZE. Each pane is an idle claude process; 4 is already a
+// lot of resident memory on a small host (a Pi serving a family) for zero in-flight work.
+export const POOL_MAX_SIZE = 4;
+
+// A warm pane older than this is dropped on acquire rather than handed out. The periodic
+// reap tick (server.mjs) drains the pool every 15 min anyway, so this only bites when
+// that tick kept getting skipped because the TUI path was never idle. Guards against
+// handing out a pane whose `claude` has been sitting so long it may have drifted
+// (auto-compaction prompts, an idle-disconnect banner, an expired in-pane token).
+export const POOL_MAX_AGE_MS = 10 * 60 * 1000;
+
+// Clamp the operator-supplied size into [0, POOL_MAX_SIZE]. A garbage value disables the
+// pool rather than guessing — an unparseable size must never silently boot 4 processes.
+export function resolvePoolSize(raw) {
+  const n = parseInt(raw, 10);
+  if (!Number.isFinite(n) || n <= 0) return 0;
+  return Math.min(n, POOL_MAX_SIZE);
+}
+
+export class TuiPanePool {
+  // size:       target number of warm panes (0 = disabled).
+  // maxAgeMs:   per-pane TTL (see POOL_MAX_AGE_MS).
+  // mintPane:   () => ({ sessionId, name }) — mints the identity of the NEXT pane. The POOL,
+  //             not the boot function, owns this: the tmux session springs into existence the
+  //             instant bootPane starts, so the pool must already know its NAME (see
+  //             _bootingPane below). Deriving the name from the sessionId also makes `tmux ls`
+  //             correlate to the transcript file.
+  // bootPane:   async (model, {sessionId, name}) => { name, sessionId, model, bootedAt } —
+  //             boots ONE pane under exactly that identity and resolves only once it is
+  //             input-ready; throws if it never becomes ready.
+  // killPane:   (name) => void — tmux kill-session. MUST be synchronous (see drain).
+  // paneHealthy:(name) => bool — pane still exists AND is still at its input bar.
+  constructor({ size, maxAgeMs = POOL_MAX_AGE_MS, mintPane, bootPane, killPane, paneHealthy, now = Date.now, log = () => {} }) {
+    this.size = Math.max(0, Math.min(parseInt(size, 10) || 0, POOL_MAX_SIZE));
+    // Fail fast at CONSTRUCTION, not at request time. refill() is called synchronously from
+    // the request path (runTuiTurn), so a missing collaborator would otherwise surface as a
+    // 500 on a live request instead of a loud error at boot.
+    if (this.size > 0) {
+      for (const [k, fn] of [["mintPane", mintPane], ["bootPane", bootPane], ["killPane", killPane], ["paneHealthy", paneHealthy]]) {
+        if (typeof fn !== "function") throw new TypeError(`TuiPanePool: ${k} must be a function`);
+      }
+    }
+    this.maxAgeMs = maxAgeMs;
+    this._mintPane = mintPane;
+    this._bootPane = bootPane;
+    this._killPane = killPane;
+    this._paneHealthy = paneHealthy;
+    this._now = now;
+    this._log = log;
+
+    this._panes = [];        // warm, available panes: { name, sessionId, model, bootedAt }
+    // The pane currently BOOTING, BY NAME ({sessionId, name, model}) — or null.
+    //
+    // WHY A NAME AND NOT A COUNT (this is a fixed bug, don't regress it): bootTuiPane creates
+    // the tmux session SYNCHRONOUSLY and only THEN waits up to POOL_BOOT_MS (20 s) for the
+    // input bar. So for up to 20 s there is a LIVE pooled tmux session. When the pool tracked
+    // only a count, it could not NAME that session, so:
+    //   - liveNames() could not spare it and the periodic reap sweep KILLED it (and
+    //     kill-server'd on top), leaving the pool empty with nothing scheduled and firing the
+    //     very tui_pool_boot_failed WARN operators are told to alert on; and
+    //   - drain() could not kill it, so on shutdown it ORPHANED a live authenticated `claude`
+    //     (the boot's .then that was supposed to clean up never runs — gracefulShutdown calls
+    //     process.exit in the same tick).
+    // Both are fixed by holding the identity here, before the session exists.
+    this._bootingPane = null;
+    // Generation counter. Bumped whenever an in-flight boot is CANCELLED (drain / model
+    // switch). A boot compares the generation it started under against the current one:
+    // if they differ, its pane was already killed by us and its settle is inert — in
+    // particular a rejection is a CANCELLATION, not an operator-visible boot failure.
+    this._gen = 0;
+    this._paused = false;    // true while drained; refill() is a no-op until resume()
+    this.warmModel = null;   // the model the pool currently warms — learned from traffic (see acquire)
+
+    this.hits = 0;           // requests served by a warm pane
+    this.misses = 0;         // requests that fell back to the cold path
+    this.boots = 0;          // panes successfully pre-booted
+    this.bootFailures = 0;   // pre-boots that genuinely never reached the input bar
+    this.cancelled = 0;      // in-flight boots WE killed (drain / model switch) — not failures
+    this.dropped = 0;        // panes discarded unused (unhealthy / expired / wrong model / drained /
+                             // cancelled — a cancelled in-flight boot also lands here via _drop)
+  }
+
+  get enabled() { return this.size > 0; }
+  get warm() { return this._panes.length; }
+  get booting() { return this._bootingPane ? 1 : 0; }
+
+  // The reaper's spare set: the EXACT names of every pane the pool currently owns and has NOT
+  // handed out — the warm ones AND the one currently booting (whose tmux session is already
+  // live; see _bootingPane). See the POOL/REAPER INVARIANT in lib/tui/session.mjs.
+  // Fail-safe by construction: a pane leaves this set the instant it is acquired, dropped, or
+  // cancelled, and if the pool is empty (or the process restarted) the set is empty — so an
+  // orphaned pooled pane looks exactly like any other stale session and IS reaped.
+  liveNames() {
+    const names = new Set(this._panes.map((p) => p.name));
+    if (this._bootingPane) names.add(this._bootingPane.name);
+    return names;
+  }
+
+  // Take a warm pane for `model`, or null (caller must fall back to the cold path — a MISS
+  // is always safe, never an error). Synchronous: paneHealthy is a cheap tmux capture.
+  //
+  // The pool warms the MOST RECENTLY REQUESTED model (`warmModel`). There is no boot-time
+  // pre-warm and no configured model: OCP cannot know which model the next caller wants, and
+  // pre-booting a process for a model nobody asks for is pure waste. Consequence, stated
+  // plainly: the FIRST request after start (and the first after a model switch) is always a
+  // MISS. The pool pays off for the steady repeat traffic it exists to serve.
+  acquire(model) {
+    if (!this.enabled) return null;
+
+    // Retarget on a model switch: --model is fixed at spawn, so panes for another model are
+    // useless. Drop them now (they are replaced by the next refill) rather than holding
+    // processes for a model that is no longer being asked for. This includes any pane
+    // currently BOOTING for the old model — its tmux session already exists, so leaving it to
+    // die on resolve would both hold a useless process and block the next refill (one boot at
+    // a time) for up to POOL_BOOT_MS.
+    if (model !== this.warmModel) {
+      for (const p of this._panes) { this._drop(p, "model_switch"); }
+      this._panes = [];
+      this._cancelBooting("model_switch");
+      this.warmModel = model;
+    }
+
+    while (this._panes.length) {
+      const p = this._panes.shift();
+      if (this._now() - p.bootedAt > this.maxAgeMs) { this._drop(p, "expired"); continue; }
+      if (!this._paneHealthy(p.name)) { this._drop(p, "unhealthy"); continue; }
+      this.hits++;
+      return p; // caller OWNS it now: it is out of the registry (so out of the spare set),
+                // and the caller's finally MUST kill it. Single-use — never returned here.
+    }
+    this.misses++;
+    return null;
+  }
+
+  // Bring the pool back up to `size` warm panes for `warmModel`. Fire-and-forget: never
+  // awaited on the request path and never throws into it.
+  //
+  // SLOT ACCOUNTING: a refill boot deliberately does NOT take a TuiSemaphore slot. Those
+  // slots bound concurrent *turns* (each up to the 120 s wallclock) and belong to real
+  // requests; charging a background pre-boot against them would let the pool starve the
+  // traffic it exists to speed up. It cannot leak a slot either, because it never holds one.
+  //
+  // SERIALIZED, ONE BOOT AT A TIME (and re-kicked on success until the pool is at target).
+  // An earlier version launched all `want` boots at once; live at size=2 that put two cold
+  // `claude` boots plus an in-flight turn on the CPU together, and a refill overran even the
+  // generous pool readiness cap (tui_pool_boot_failed). Booting sequentially keeps each boot
+  // near its uncontended ~1.2 s, bounds the CPU burst the pool can cause, and still has the
+  // replacement pane warm long before the next request arrives.
+  //
+  // A genuinely FAILED boot deliberately does NOT re-kick the chain — that is the backoff. A
+  // persistently failing boot (bad claude binary, no auth) would otherwise spin, respawning
+  // forever. The next natural trigger (the following request's refill, or the reap tick's
+  // resume) retries it. A CANCELLED boot is different: we killed it on purpose, nothing is
+  // wrong, and resume() is expected to start a fresh one immediately.
+  refill() {
+    if (!this.enabled || this._paused || !this.warmModel) return;
+    if (this._bootingPane) return;                       // one boot in flight at a time
+    if (this._panes.length >= this.size) return;         // already at target
+
+    const model = this.warmModel;
+    const gen = this._gen;
+    // Mint the identity BEFORE booting: bootPane creates the tmux session synchronously, so
+    // the pool must be able to name (and therefore spare, and kill) it from this moment on.
+    const ident = this._mintPane();
+    this._bootingPane = { ...ident, model };
+    let enlisted = false;
+    Promise.resolve()
+      .then(() => this._bootPane(model, ident))
+      .then((pane) => {
+        // The world may have moved while we booted. If our generation was cancelled, kill the
+        // pane here rather than ASSUMING _cancelBooting already did.
+        //
+        // Why not just `return`: _cancelBooting kills by name, but the tmux session only EXISTS
+        // once _bootPane has actually run — and _bootPane is queued on a microtask (above). A
+        // caller that does refill() and then drain() in the SAME synchronous block would have
+        // _cancelBooting find nothing to kill (a no-op), bump the generation, and then this
+        // microtask would create the session, boot it fine, and — under a bare `return` — walk
+        // away from a LIVE authenticated `claude` that nothing owns. That is M1b in a new costume.
+        // No current call site does that, so this is defense-in-depth, not a live bug — but ADR
+        // 0008 and the reap-tick comment in server.mjs both explicitly contemplate a boot-time
+        // pre-warm, which is exactly the shape that would reach it.
+        //
+        // Killing an already-dead session is a harmless no-op (_drop swallows it), so this is
+        // idempotent whether or not _cancelBooting got there first.
+        if (gen !== this._gen) { this._drop(pane, "cancelled_late"); return; }
+        // Otherwise: still possible the pool filled or retargeted without a cancellation.
+        if (this._paused || model !== this.warmModel || this._panes.length >= this.size) {
+          this._drop(pane, "stale_boot");
+          return;
+        }
+        this._panes.push(pane);
+        this.boots++;
+        enlisted = true;
+      })
+      .catch((e) => {
+        // A rejection from a CANCELLED generation is not a fault: it is almost always
+        // "tui_pane_not_ready", thrown because WE killed the pane out from under the boot.
+        // Counting it as a bootFailure would fire the exact WARN operators are told to alert
+        // on, for a completely healthy drain. Stay silent — _cancelBooting already counted
+        // this as a cancellation, so do NOT count it again here.
+        if (gen !== this._gen) return;
+        this.bootFailures++;
+        this._log("warn", "tui_pool_boot_failed", { model, error: e && e.message });
+      })
+      .finally(() => {
+        // ONLY the current generation's boot owns the booting slot. A stale settle must not
+        // clear a slot that a newer boot (started by resume()) already holds.
+        if (gen === this._gen) this._bootingPane = null;
+        if (enlisted) this.refill(); // continue toward target, still one at a time
+      });
+  }
+
+  // Kill the in-flight boot's pane, SYNCHRONOUSLY, and invalidate its generation. Returns 1
+  // if there was one, else 0. The tmux session already exists (bootPane created it before it
+  // started waiting for readiness), so this is a real kill, not a cancellation flag.
+  _cancelBooting(reason) {
+    if (!this._bootingPane) return 0;
+    this._gen++;                              // the in-flight boot's settle is now inert
+    this._drop(this._bootingPane, reason);    // synchronous kill-session
+    this._bootingPane = null;
+    this.cancelled++;
+    return 1;
+  }
+
+  // Kill every pane the pool owns — warm AND currently booting — and stop refilling. Returns
+  // how many were killed.
+  //
+  // Called (a) before the periodic reap sweep — reapStaleTuiSessions can only reap defunct
+  // `claude` zombies via kill-server, and kill-server is suppressed while any live pooled pane
+  // exists (including a booting one), so without this drain the pool would permanently disable
+  // zombie reaping; and (b) on graceful shutdown, so no pane outlives the process as an orphan.
+  //
+  // EVERY KILL HERE IS SYNCHRONOUS, and that is load-bearing. It is NOT safe to leave the
+  // booting pane to clean itself up on resolve: gracefulShutdown calls process.exit() in the
+  // same tick as this drain (TUI panes are children of the tmux SERVER, not of node, so
+  // node's activeProcesses set is empty on a TUI host and the "wait for children" path exits
+  // immediately). A .then()/.catch() scheduled here would never run, and the pane would
+  // survive as an orphaned, authenticated, idle `claude`.
+  drain() {
+    this._paused = true;
+    let n = this._panes.length;
+    for (const p of this._panes) this._drop(p, "drain");
+    this._panes = [];
+    n += this._cancelBooting("drain_booting");
+    return n;
+  }
+
+  // Undo drain() and start refilling again. Because drain() CANCELLED the in-flight boot
+  // (rather than leaving it pending), the booting slot is free and this really does start a
+  // fresh boot — the pool is never left empty with nothing scheduled.
+  resume() {
+    this._paused = false;
+    this.refill();
+  }
+
+  // /health surface (additive).
+  stats() {
+    return {
+      size: this.size,
+      warm: this._panes.length,
+      booting: this.booting,
+      model: this.warmModel,
+      hits: this.hits,
+      misses: this.misses,
+      boots: this.boots,
+      bootFailures: this.bootFailures,
+      cancelled: this.cancelled,
+      dropped: this.dropped,
+    };
+  }
+
+  _drop(pane, reason) {
+    this.dropped++;
+    try { this._killPane(pane.name); } catch { /* already gone */ }
+    // F5: every drop path (expired / unhealthy / model_switch / drain / cancelled_late /
+    // stale_boot) ends up here, and the reap tick drains the WHOLE pool on every tick — so
+    // without this, every warm pane's sink orphans in streamDir with no GC path (killPane only
+    // reaches the tmux session, never the pane's OWN files). Best-effort: pane.streamFile is
+    // undefined for a still-booting identity (the sink path is only known once bootPane
+    // resolves) and rmSync(force:true) is already a no-op on a missing file, so this never
+    // throws into the reaper regardless of which drop path got here.
+    if (pane.streamFile) {
+      try { rmSync(pane.streamFile, { force: true }); } catch { /* best-effort GC */ }
+    }
+    this._log("info", "tui_pool_pane_dropped", { name: pane.name, reason });
+  }
+}
@@ -139,7 +139,40 @@ export function recordTuiEntrypoint(tuiStats, observed, expectedMode = "cli") {
 // Build the additive /health `tui` block (ADR 0007 PR-B amendment). Pure: given the
 // config + live counters, returns the exact object embedded in /health. New fields only —
 // behaviour-preserving for existing /health consumers (grandfathered B.2 under ADR 0006).
-export function buildTuiHealthBlock({ enabled, entrypointMode, maxConcurrent }, tuiStats, semaphore) {
+//
+// `pool` (optional, warm pane pool — lib/tui/pool.mjs): a TuiPanePool, or null/undefined
+// when the pool is off (the default). Reported as `pool: null` when off so the block's
+// shape stays stable, and as the pool's stats (size / warm / hits / misses / …) when on —
+// the operator's window onto both the hit rate and the standing idle-process cost.
+//
+// Streaming fields (backlog #2, OCP_TUI_STREAM) are ADDITIVE too:
+//   streamEnabled       — is real (MessageDisplay-hook) SSE streaming on for TUI turns?
+//   streamTurns         — streamed turns ATTEMPTED, counted before the truncation/auth-banner
+//                         gates run (F6) — so a turn REFUSED by those gates still shows up
+//                         here, which is exactly the turn an operator most wants visible.
+//                         Counting only turns that survived the gates would silently exclude
+//                         a turn's worst-case outcome from its own denominator.
+//   streamDeltas        — MessageDisplay hook fires OBSERVED, including held-back ones (F6) —
+//                         NOT only the ones forwarded to a client. This is what makes
+//                         streamZeroDeltaTurns meaningful: a turn can have streamDeltas
+//                         incrementing while still emitting nothing to the client (fully held
+//                         back, e.g. a short answer), which is healthy, vs. a hook that fired
+//                         zero times at all, which is not (see streamZeroDeltaTurns).
+//   streamTopUps        — turns where the delta stream was a safe PREFIX of the transcript but
+//                         not equal to it; OCP topped up from the transcript and served T.
+//                         Benign but worth watching — a persistent rate means the hook is
+//                         losing fires.
+//   streamDivergences   — turns REFUSED because emitted bytes were not a prefix of the
+//                         transcript. THE field to alert on for CORRECTNESS: it means the hook
+//                         and the transcript disagreed and OCP chose to fail rather than serve
+//                         unverifiable text.
+//   streamZeroDeltaTurns — streamed turns where the hook fired ZERO times (F7). THE field to
+//                         alert on for AVAILABILITY: streamTopUps climbing is one fire dropped
+//                         here and there (benign); this climbing means the hook is not firing
+//                         AT ALL — e.g. `--settings` silently stopped registering it (a claude
+//                         version bump), or F3's truncated-script failure mode — and every
+//                         streamed turn is quietly degrading to fully-buffered with no error.
+export function buildTuiHealthBlock({ enabled, entrypointMode, maxConcurrent, streamEnabled = false }, tuiStats, semaphore, pool = null) {
  return {
    enabled,
    entrypointMode,                                  // cli | auto | off
@@ -148,5 +181,12 @@ export function buildTuiHealthBlock({ enabled, entrypointMode, maxConcurrent },
    inflight: semaphore.inflight,                    // current concurrent TUI turns
    queued: semaphore.queued,                        // turns waiting for a slot
    maxConcurrent,
+    pool: pool ? pool.stats() : null,                // warm pane pool, or null when disabled
+    streamEnabled,
+    streamTurns: tuiStats.streamTurns ?? 0,
+    streamDeltas: tuiStats.streamDeltas ?? 0,
+    streamTopUps: tuiStats.streamTopUps ?? 0,
+    streamDivergences: tuiStats.streamDivergences ?? 0,
+    streamZeroDeltaTurns: tuiStats.streamZeroDeltaTurns ?? 0,
  };
 }
@@ -14,6 +14,7 @@ import { mkdtempSync, writeFileSync, readFileSync, mkdirSync, existsSync, rmSync
 import { tmpdir } from "node:os";
 import { randomUUID } from "node:crypto";
 import { readTuiTranscript } from "./transcript.mjs";
+import { prepareStreamHook, streamFilePath, parseDeltaChunk } from "./stream.mjs";

 // F7 fix (audit finding, LOW): the prefix used to be a bare, host-wide constant
 // ("ocp-tui-"), so a SECOND OCP instance on the same host (e.g. a temporary
@@ -73,6 +74,36 @@ const defaultTmux = (args, opts = {}) =>
 // `port` (required) is this instance's own listen port (server.mjs's PORT / lib/constants.mjs
 // DEFAULT_PORT resolution) — the SPOT for "which sessions are ours."
 //
+// ── POOL/REAPER INVARIANT (warm pane pool — lib/tui/pool.mjs) ───────────────────────────
+// A warm pooled pane is one of OUR OWN `ocp-tui-<port>-*` sessions that is ALIVE AND IDLE
+// BY DESIGN — and the periodic sweep runs precisely when the instance is idle, i.e. exactly
+// when the pool is full. Without an exemption the sweep would kill every warm pane on every
+// tick (and kill-server on top). The exemption is `spare`: a set of EXACT session names the
+// caller declares live. Three properties, all load-bearing:
+//
+//   1. A LIVE POOLED PANE IS NEVER REAPED — INCLUDING ONE THAT IS STILL BOOTING. It is in
+//      `spare` (the pool's live registry), so it is skipped by name. The booting case is not
+//      a footnote, it is the one that bit us: bootTuiPane creates the tmux session
+//      SYNCHRONOUSLY and only then waits up to POOL_BOOT_MS for the input bar, so a pooled
+//      session can be live for ~20 s before its boot resolves. The pool therefore mints the
+//      pane's NAME up front and holds it in `_bootingPane`, so liveNames() can name — and
+//      spare — a session whose boot has not finished. (An earlier version tracked only a
+//      COUNT of in-flight boots; the sweep could not name that session and killed it.)
+//   2. A LEAKED/ORPHANED POOLED PANE IS STILL REAPED. Membership is by EXACT NAME from a
+//      live in-memory registry — NOT by "looks pooled" (name shape). A pane the pool no
+//      longer owns (handed out, dropped, cancelled, or left behind by a previous process
+//      generation — whose registry died with it) is absent from `spare` and is killed like
+//      any other stale session. Fail-safe: forgetting to pass `spare` reaps MORE, never less.
+//   3. KILL-SERVER NEVER KILLS A LIVE POOL PANE. A spared session suppresses kill-server
+//      exactly as a foreign session does (it is a live child of the tmux server). The
+//      consequence — that a permanently-full pool would permanently disable the defunct-
+//      zombie reaping that ONLY kill-server can do — is resolved in server.mjs by DRAINING
+//      the pool immediately before the sweep, so `spare` is empty on the normal tick and
+//      kill-server still fires. `spare` is the belt-and-braces: a reap call site that
+//      forgets to drain still cannot kill a live pane.
+//
+// `spare` (default: none) — iterable of session names, or a Set. Ignored when the pool is off.
+//
 // `includeLegacy` (default false): when true, sessions matching the exact OLD bare-prefix
 // shape (LEGACY_SESSION_NAME_RE) are ALSO treated as ours for kill-session purposes. This is
 // the boot-time legacy migration: an operator upgrading past this fix could otherwise be left
@@ -88,14 +119,20 @@ const defaultTmux = (args, opts = {}) =>
 // same class of residual risk the audit finding itself accepts ("no live instance of the new
 // version creates them"); this PR does not regress that scenario, it only removes the far
 // more common same-version collision (the actual F7 finding).
-export function reapStaleTuiSessions({ tmux = defaultTmux, port, includeLegacy = false } = {}) {
+export function reapStaleTuiSessions({ tmux = defaultTmux, port, includeLegacy = false, spare = null } = {}) {
  const r = tmux(["list-sessions", "-F", "#{session_name}"]);
  if (!r || r.status !== 0) return 0; // no tmux server / no sessions
  const names = String(r.stdout || "").split("\n").map((s) => s.trim()).filter(Boolean);
  const ownPrefix = sessionPrefixForPort(port);
+  const spared = spare instanceof Set ? spare : new Set(spare || []);
  let killed = 0;
  let othersRemain = false;
+  let sparedLive = 0;
  for (const name of names) {
+    // Property 1+2: exemption is by EXACT NAME from the pool's live registry. A pooled-
+    // LOOKING name that is not in the registry is an orphan and falls through to the
+    // normal kill path below.
+    if (spared.has(name)) { sparedLive++; continue; }
    const isOwn = name.startsWith(ownPrefix);
    const isLegacyOwn = includeLegacy && LEGACY_SESSION_NAME_RE.test(name);
    if (isOwn || isLegacyOwn) {
@@ -109,7 +146,11 @@ export function reapStaleTuiSessions({ tmux = defaultTmux, port, includeLegacy =
  // Reap defunct `claude` zombies: safe ONLY when the server is now ours-only/empty.
  // kill-server is what actually reaps (server exit reparents survivors to init); a
  // per-session kill cannot, since node is not the zombies' parent.
-  if (!othersRemain) {
+  //
+  // Property 3: a SPARED session is a live child of this tmux server, so kill-server would
+  // kill it — it therefore suppresses kill-server exactly as a foreign session does. On the
+  // normal sweep the pool is drained first, so sparedLive is 0 and kill-server still fires.
+  if (!othersRemain && sparedLive === 0) {
    tmux(["kill-server"]);
  }
  return killed;
@@ -119,8 +160,18 @@ export function reapStaleTuiSessions({ tmux = defaultTmux, port, includeLegacy =

 // Boot + paste-settle timing. Conservative defaults validated on PI231; env-tunable.
 const BOOT_MS         = parseInt(process.env.OCP_TUI_BOOT_MS  || "4000", 10);   // max wait for input-ready
+// Readiness cap for a POOL pre-boot. Deliberately far more generous than BOOT_MS: BOOT_MS is
+// tight because a client is blocked on it, whereas a warm-pane boot happens in the background
+// with nobody waiting. Observed live at size=2: a refill booting alongside an in-flight turn
+// exceeded 4000 ms and was discarded (tui_pool_boot_failed), quietly costing hit rate for a
+// pane that was merely slow, not broken. Scales with OCP_TUI_BOOT_MS if an operator raises it.
+export const POOL_BOOT_MS = BOOT_MS * 5;
 const READY_POLL_MS   = parseInt(process.env.OCP_TUI_READY_POLL_MS || "400", 10); // readiness / paste-verify poll interval
 const PASTE_VERIFY_MS = parseInt(process.env.OCP_TUI_PASTE_VERIFY_MS || "5000", 10); // max wait for pasted prompt to render
+// Hook-sink drain interval when streaming. 100ms: the hook fires at BLOCK granularity
+// (~5-7 fires per answer, seconds apart), so a finer poll buys nothing and a coarser one
+// would add visible lag to the first delta. Cheap — one readFileSync of a small file.
+const STREAM_POLL_MS  = parseInt(process.env.OCP_TUI_STREAM_POLL_MS || "100", 10);

 const sleep = (ms) => new Promise((r) => setTimeout(r, ms));

@@ -302,7 +353,25 @@ export function prepareTuiHome(realHome, tuiHome, cwd, { envTokenMode = false }
 // A-PATH ONLY: built-in tools are left enabled (acceptable single-user). Deployment B
 // (guest keys) MUST additionally pass --tools "" per spec §5.2(2) as the credential
 // wall before this argv is reachable for owner_tier=guest — guard that in PR-3 wiring.
-export function buildTuiCmd(claudeBin, model, sessionId, ehome, entrypointMode) {
+//
+// `stream` (optional, OCP_TUI_STREAM): { file, settings } — when present, the pane gets
+//   (a) OCP_TUI_STREAM_FILE in its env  — read by the static MessageDisplay hook script to
+//       decide WHERE to append this pane's deltas. Delivered as env (not baked into the
+//       settings file) so the settings file stays STATIC and a pre-booted warm pane works.
+//       Verified live: a claude hook inherits the pane's environment.
+//   (b) --settings <file>               — registers the MessageDisplay hook.
+//   VERIFIED LIVE (claude 2.1.207, this host) before shipping, because both were spawn-level
+//   risks:
+//     - the startup banner is UNCHANGED with --settings: "Sonnet 4.6 with low effort ·
+//       Claude Max" (subscription pool). --settings is NOT a --bare-class flag — it does not
+//       silently drop the subscription pool. Transcript entrypoint stayed "cli".
+//     - --settings MERGES into the settings hierarchy, it does NOT clobber <HOME>/.claude/
+//       settings.json: with --settings passed, the user-level settings.json's `env` block was
+//       still applied to the hook's environment. So the isolated-HOME settings story the TUI
+//       already relies on (permissions / additionalDirectories — see prepareTuiHome and the
+//       OCP_TUI_FULL_TOOLS note above) survives intact.
+//   When absent, the argv is byte-for-byte the pre-streaming argv.
+export function buildTuiCmd(claudeBin, model, sessionId, ehome, entrypointMode, stream = null) {
  // Deliver claude's env via an `env` prefix on the PANE COMMAND — tmux does NOT forward the
  // spawning process's environment to the pane, and `new-session -e` needs tmux ≥3.2 (the cloud
  // host runs 2.7), so this is the only portable, reliable mechanism (verified live 2026-06-01:
@@ -346,6 +415,8 @@ export function buildTuiCmd(claudeBin, model, sessionId, ehome, entrypointMode)
  if (process.env.CLAUDE_CODE_OAUTH_TOKEN) {
    sets.push(`CLAUDE_CODE_OAUTH_TOKEN=${shq(process.env.CLAUDE_CODE_OAUTH_TOKEN)}`);
  }
+  // Streaming sink: the pane's own per-session delta file (see the `stream` note above).
+  if (stream && stream.file) sets.push(`OCP_TUI_STREAM_FILE=${shq(stream.file)}`);
  const unset = ["CLAUDECODE", "ANTHROPIC_API_KEY", "ANTHROPIC_BASE_URL", "ANTHROPIC_AUTH_TOKEN"];
  if (entrypointMode === "cli") sets.push("CLAUDE_CODE_ENTRYPOINT=cli");
  else if (entrypointMode === "auto") unset.push("CLAUDE_CODE_ENTRYPOINT"); // let claude self-classify via TTY
@@ -377,47 +448,107 @@ export function buildTuiCmd(claudeBin, model, sessionId, ehome, entrypointMode)
  } else {
    toolArgs = ["--strict-mcp-config", "--disallowedTools", shq("mcp__*")];
  }
+
+  // Effort: pass --effort EXPLICITLY. Without it, the pane's claude inherits a
+  // HOME-dependent effortLevel — real-home mode inherits the operator's
+  // ~/.claude/settings.json (whatever they set for their own interactive use),
+  // env-token scratch mode inherits claude's built-in default (prepareTuiHome never
+  // writes effortLevel) — so latency silently depends on which HOME mode
+  // resolveTuiHome() picked AND on an unrelated operator setting. Pinning it here
+  // removes both. Measured (docs/plans/2026-07-13-tui-latency): explicit low cuts
+  // direct-spawn TTFT p50 10.35s → 6.17s (−40%) and collapses the spread ~15×;
+  // banner-verified to stay on the subscription pool (`· Claude Max`).
+  // OCP_TUI_EFFORT=inherit restores the pre-flag argv byte-for-byte (no --effort).
+  // An unknown value falls back to the default rather than reaching claude's argv:
+  // a typo'd --effort value must not risk a spawn-time usage error in the pane.
+  const EFFORT_LEVELS = ["low", "medium", "high", "xhigh", "max"]; // claude 2.1.207 --help
+  const effortRaw = (process.env.OCP_TUI_EFFORT || "low").trim().toLowerCase();
+  let effortArgs;
+  if (effortRaw === "inherit") {
+    effortArgs = [];
+  } else if (EFFORT_LEVELS.includes(effortRaw)) {
+    effortArgs = ["--effort", effortRaw];
+  } else {
+    console.error(`[tui] invalid OCP_TUI_EFFORT=${JSON.stringify(process.env.OCP_TUI_EFFORT)}; using "low" (valid: ${EFFORT_LEVELS.join("|")}, or "inherit" to omit the flag)`);
+    effortArgs = ["--effort", "low"];
+  }
+
+  // --settings registers the MessageDisplay hook. Omitted entirely when streaming is off,
+  // so the OFF argv is byte-for-byte the pre-streaming argv.
+  const settingsArgs = stream && stream.settings ? ["--settings", shq(stream.settings)] : [];
+
  return [
    envPrefix,
    shq(claudeBin),
    "--model",          shq(model),
    "--session-id",     sessionId,
    ...toolArgs,
+    ...effortArgs,
+    ...settingsArgs,
  ].join(" ");
 }

-// Full per-request TUI lifecycle:
-//   1. Pre-trust the scratch cwd (no trust dialog will appear).
-//   2. Write prompt to a 0600 temp file (no shell injection from prompt content).
-//   3. Boot an interactive `claude` in a fresh tmux session in the scratch cwd; poll
-//      capture-pane until the `? for shortcuts` input bar appears (readiness-poll
-//      replaces the old blind boot sleep). BOOT_MS is the max wait, not a fixed delay.
-//   4. Paste the prompt via tmux load-buffer + paste-buffer -p (bracketed paste) —
-//      reliable for large multi-line prompts where send-keys -l is not (issue #130).
-//      Poll-verify the prompt landed in the input (placeholder gone / [Pasted text]);
-//      fast-fail with tui_paste_not_landed if it never lands (prevents the 120s
-//      wallclock "stuck typing" hang). Then submit with a SEPARATE Enter key event.
-//   5. Block on the native JSONL transcript (located by session-id) until terminal
-//      marker or wall-clock cap.
-//   6. Always teardown: kill session + rm temp dir (even on throw).
-// Returns { text, entrypoint } from readTuiTranscript (entrypoint is the billing-pool
-// classifier, e.g. "cli", or null if the transcript did not include a turn_duration).
-export async function runTuiTurn({
-  prompt,
-  model,
-  claudeBin,
-  home,
-  realHome,
-  cwd,
-  port,
-  wallclockMs = 120000,
-  entrypointMode = "cli",
-  tmux = defaultTmux,
+// Is a pane alive AND still sitting at its input bar? Used by the warm pool to decide,
+// at hand-out time, whether a pre-booted pane is still usable (a dead/degraded pane must
+// become a MISS → cold path, never a hung turn). capture-pane exits non-zero when the
+// session no longer exists, so this covers "pane gone" and "pane not ready" in one call.
+export function tuiPaneHealthy(tmux, tmuxName) {
+  const r = tmux(["capture-pane", "-p", "-t", tmuxName]);
+  if (!r || r.status !== 0 || typeof r.stdout !== "string") return false;
+  return tuiInputReady(r.stdout);
+}
+
+// Pool pane names carry a "p" marker after the port-scoped prefix:
+//   turn pane: ocp-tui-<port>-<8hex>      (unchanged)
+//   pool pane: ocp-tui-<port>-p<8hex>
+// Purely for operator legibility (`tmux ls` shows which panes are warm). It is NOT the
+// reaper's exemption mechanism — that is the exact-name spare set (see the POOL/REAPER
+// INVARIANT above), so a pooled-LOOKING orphan is still reaped. Both shapes start with
+// sessionPrefixForPort(port), so both remain reapable as "ours", and neither can match
+// LEGACY_SESSION_NAME_RE.
+export function poolPaneName(port, sessionId) {
+  return sessionPrefixForPort(port) + "p" + sessionId.slice(0, 8);
+}
+
+// Boot ONE interactive `claude` pane and wait for its input bar. Shared by the cold
+// request path (runTuiTurn) and the warm pool (lib/tui/pool.mjs) so a pooled pane is
+// spawned with byte-for-byte the same argv, HOME, cwd and trust preparation as a
+// cold-booted one — the pool must not become a second, drifting spawn path.
+//
+// Each pane gets its OWN fresh randomUUID() --session-id, fixed at boot. That is what
+// keeps a pooled pane single-use-safe: its transcript holds exactly one exchange.
+//
+// requireReady: the cold path tolerates a readiness timeout (it falls through and lets
+// the paste-verify decide — pre-existing behaviour, unchanged). The POOL sets it, because
+// a pane that never reached its input bar is worthless as a warm pane and must not be
+// enlisted: throw, let the pool count a bootFailure, and leave the request path to
+// cold-boot as usual.
+// bootMs: max wait for the input bar. Defaults to BOOT_MS (the REQUEST path's cap, which is
+// deliberately tight — a client is blocked on it). The POOL passes POOL_BOOT_MS instead: a
+// background pre-boot has nobody waiting on it, and capping it at the request-path's 4 s
+// made real refills fail (observed live: a refill booting alongside an in-flight turn took
+// >4 s and was discarded, silently lowering the hit rate). Slow != broken for a pre-boot.
+// `sessionId` / `name` (both optional): the caller may supply the pane's identity instead of
+// letting bootTuiPane mint it. The POOL does, because it must know the tmux session's NAME
+// before this function runs — the session is created synchronously below, well before the
+// readiness wait returns, so a pool that only learned the name on resolve could neither spare
+// the session from the reaper nor kill it on shutdown. Supplying BOTH also keeps the name's
+// hex suffix equal to the session-id's, so `tmux ls` correlates to the transcript file.
+// `streamDir` (optional, OCP_TUI_STREAM): install claude's MessageDisplay hook on this pane.
+// Done HERE, at boot — not at turn time — and that is the whole reason streaming survives the
+// WARM POOL: the hook script + settings file are STATIC (one pair per streamDir), and the only
+// per-turn thing, the sink path, is derived from the pane's own --session-id, which is fixed
+// right here. So a pre-booted pane already carries its hook and its own sink and streams exactly
+// like a cold-booted one; nothing request-specific is ever baked into the spawn.
+export async function bootTuiPane({
+  model, claudeBin, home, realHome, cwd, port, entrypointMode = "cli",
+  tmux = defaultTmux, sessionId = null, name = null, requireReady = false, bootMs = BOOT_MS,
+  streamDir = null,
 }) {
-  const sessionId = randomUUID();
+  const sid = sessionId || randomUUID();
  // Port-scoped session name (F7 fix) — see sessionPrefixForPort / reapStaleTuiSessions
  // for why this instance's own listen port is the namespace discriminator.
-  const tmuxName  = sessionPrefixForPort(port) + sessionId.slice(0, 8);
+  const tmuxName  = name || (sessionPrefixForPort(port) + sid.slice(0, 8));
  const ehome     = home || process.env.HOME;       // HOME claude runs under (scratch or real)
  const rhome     = realHome || process.env.HOME;   // real home (OAuth + onboarded config source)

@@ -434,10 +565,14 @@ export async function runTuiTurn({
  if (!existsSync(cwd)) mkdirSync(cwd, { recursive: true });
  prepareTuiHome(rhome, ehome, cwd, { envTokenMode });

-  // Write prompt to a temp file (mode 0600) so the content never touches argv.
-  const tmpDir     = mkdtempSync(`${tmpdir()}/ocp-tui-`);
-  const promptFile = `${tmpDir}/prompt.txt`;
-  writeFileSync(promptFile, prompt, { mode: 0o600 });
+  // Streaming sink for THIS pane (see the streamDir note above). rmSync first so a
+  // re-used session-id can never replay a previous turn's deltas.
+  let streamFile = null, streamSettings = null;
+  if (streamDir) {
+    streamFile     = streamFilePath(streamDir, sid);
+    streamSettings = prepareStreamHook(streamDir);
+    try { rmSync(streamFile, { force: true }); } catch { /* start from a fresh sink */ }
+  }

  // Minimal env for spawnSync (tmux itself). The pane's claude env comes exclusively
  // from the `env` prefix string built inside buildTuiCmd — tmux does NOT forward the
@@ -445,31 +580,147 @@ export async function runTuiTurn({
  const env = { ...process.env };
  env.HOME = ehome; // tmux needs HOME; all claude-specific vars go via buildTuiCmd prefix

+  // Boot the interactive session inside tmux, rooted at the scratch cwd.
+  // Capture the result: if tmux new-session fails (status !== 0) there is no PTY, no
+  // interactive spawn — abort BEFORE the boot wait rather than paste into a non-existent
+  // session or issue a billing request without a verified interactive context.
+  const spawnResult = tmux(
+    ["new-session", "-d", "-s", tmuxName, "-x", "220", "-y", "50", "-c", cwd,
+     buildTuiCmd(claudeBin, model, sid, ehome, entrypointMode,
+                 streamFile ? { file: streamFile, settings: streamSettings } : null)],
+    { env },
+  );
+  if (!spawnResult || spawnResult.status !== 0) {
+    throw new Error("tui_spawn_failed: tmux session not created");
+  }
+
+  // Wait until claude's input bar is actually ready (not a blind sleep).
+  // bootMs is the MAX readiness wait, not a fixed delay.
+  const ready = await pollUntil(() => tuiInputReady(tuiCapturePane(tmux, tmuxName)),
+    { timeoutMs: bootMs, intervalMs: READY_POLL_MS });
+  if (!ready) {
+    if (requireReady) {
+      try { tmux(["kill-session", "-t", tmuxName]); } catch { /* already gone */ }
+      throw new Error("tui_pane_not_ready: input bar did not appear within " + bootMs + "ms");
+    }
+    // Cold path (pre-existing behaviour): readiness timed out; rely on paste-verify.
+    console.error("[tui] input_not_ready", tmuxName);
+  }
+  return { name: tmuxName, sessionId: sid, model, ehome, streamFile, bootedAt: Date.now() };
+}
+
+// Full per-request TUI lifecycle:
+//   1. Take a WARM pane from the pool if one is available for this model (opt-in;
+//      OCP_TUI_POOL_SIZE=0 => always null => steps 2-3 below are exactly today's path).
+//      A pooled pane is SINGLE-USE: it already carries its own fresh --session-id, it
+//      serves this one turn, and it is killed in the finally like any other pane.
+//   2. On a MISS: pre-trust the scratch cwd, boot an interactive `claude` in a fresh tmux
+//      session in the scratch cwd, poll capture-pane until the `? for shortcuts` input bar
+//      appears (bootTuiPane). BOOT_MS is the max wait, not a fixed delay.
+//   3. Write prompt to a 0600 temp file (no shell injection from prompt content).
+//   4. Paste the prompt via tmux load-buffer + paste-buffer -p (bracketed paste) —
+//      reliable for large multi-line prompts where send-keys -l is not (issue #130).
+//      Poll-verify the prompt landed in the input (placeholder gone / [Pasted text]);
+//      fast-fail with tui_paste_not_landed if it never lands (prevents the 120s
+//      wallclock "stuck typing" hang). Then submit with a SEPARATE Enter key event.
+//   5. Block on the native JSONL transcript (located by THIS pane's session-id) until
+//      terminal marker or wall-clock cap.
+//   6. Always teardown: kill session + rm temp dir (even on throw), and kick a background
+//      pool refill so the next request finds a warm pane.
+// Returns { text, entrypoint } from readTuiTranscript (entrypoint is the billing-pool
+// classifier, e.g. "cli", or null if the transcript did not include a turn_duration).
+//
+// STREAMING (OCP_TUI_STREAM, default off). Pass `onDelta` and `streamDir`, and the pane's
+// MessageDisplay hook (installed by bootTuiPane; see lib/tui/stream.mjs) appends each raw
+// delta payload to the pane's own sink. This driver polls that sink and invokes onDelta(payload)
+// per fire while the turn is still generating. A WARM pane already carries its sink from boot
+// (pane.streamFile), so the pooled and cold paths stream identically.
+//
+// `streamDir` IS PASSED TO THE COLD BOOT UNCONDITIONALLY (not gated on `onDelta`) — F4 fix. The
+// spawn argv is this project's billing-classification surface: a caller with OCP_TUI_STREAM on
+// but THIS particular request non-streaming (stream:false) must still get the SAME argv whether
+// it lands on a pool HIT or a cold-boot MISS, because a pre-booted pool pane cannot know in
+// advance whether the request it will eventually serve wants streaming — it installs the hook
+// unconditionally whenever the pool is warming at all (see server.mjs's bootPane closure). Gating
+// the cold boot's hook install on `onDelta` made a stream:false request's argv depend on whether
+// it happened to hit the pool or miss it — the exact drift this surface cannot tolerate. Whether
+// the hook is actually POLLED is a separate, correctly-scoped decision: see `streaming` below,
+// gated on onDelta && streamFile, so a non-streaming turn never reads its own sink even though
+// the hook is running.
+//
+// The transcript stays AUTHORITATIVE regardless: it is still the terminal-turn signal, still the
+// source of the returned `text`, and still the input to the caller's honesty gates. The delta
+// stream is a low-latency MIRROR of it, never a replacement, and the caller asserts the two
+// agree. With onDelta AND streamDir both omitted, nothing here changes: no poll, no hook.
+//
+// `abortSignal` (optional): aborts the transcript wait, so a client that disconnects mid-turn
+// tears the pane down NOW (the finally below) instead of holding the pane — and therefore the
+// caller's semaphore slot — until the turn or the wallclock cap ends.
+export async function runTuiTurn({
+  prompt,
+  model,
+  claudeBin,
+  home,
+  realHome,
+  cwd,
+  port,
+  wallclockMs = 120000,
+  entrypointMode = "cli",
+  tmux = defaultTmux,
+  pool = null,       // TuiPanePool | null — null (default) === today's cold-boot-only path
+  onPane = null,     // optional observer: ({ warm }) => void, for logging/metrics
+  onDelta = null,    // (payload) => void — invoked per MessageDisplay hook fire, mid-turn
+  streamDir = null,  // hook sink dir, passed to the COLD boot UNCONDITIONALLY (F4 — see above);
+                      // a warm pane brings its own, fixed at its own boot
+  abortSignal = null,
+}) {
+  // 1. Warm pane, or cold boot. A MISS is never an error — it is exactly today's path.
+  let pane = pool ? pool.acquire(model) : null;
+  const warm = !!pane;
+  // Kick the refill IMMEDIATELY (not after the turn): the replacement pane then boots
+  // CONCURRENTLY with this turn and is warm by the time the next request arrives. Also
+  // runs on a MISS — acquire() has just retargeted the pool to this model, so the miss
+  // that cold-boots today warms the pool for the next caller. Fire-and-forget; it takes
+  // no TuiSemaphore slot (see pool.refill's SLOT ACCOUNTING note).
+  if (pool) pool.refill();
+  if (onPane) { try { onPane({ warm }); } catch { /* observer must never break a turn */ } }
+  if (!pane) {
+    // streamDir passed AS-IS (not gated on onDelta) — F4: see the STREAMING comment above.
+    pane = await bootTuiPane({ model, claudeBin, home, realHome, cwd, port, entrypointMode, tmux,
+                              streamDir });
+  }
+  const tmuxName  = pane.name;
+  const sessionId = pane.sessionId;   // THIS pane's own session-id — one session, one turn
+  const ehome     = pane.ehome || home || process.env.HOME;
+
+  // Streaming state is read off the PANE, not recomputed here — a warm pane fixed its sink at
+  // boot, and a cold one just did the same above. If the pool was booted WITHOUT a streamDir
+  // while onDelta is set, streamFile is null and the turn degrades to buffered: correct, just
+  // not fast. (server.mjs wires the same streamDir into both paths so that cannot happen.)
+  const streamFile = pane.streamFile || null;
+  const streaming  = !!(onDelta && streamFile);
+  const streamCursor = { consumed: 0 };
+  let streamStopped = false;
+  let pollTimer = null;
+  // Drain every complete line appended since the last drain. Never throws into the turn: a
+  // malformed line is skipped by parseDeltaChunk, and an onDelta that throws is contained.
+  const drainDeltas = () => {
+    if (!streaming) return;
+    let text;
+    try { text = readFileSync(streamFile, "utf8"); } catch { return; } // absent until the first fire
+    const { deltas, consumed } = parseDeltaChunk(text, streamCursor.consumed);
+    streamCursor.consumed = consumed;
+    for (const d of deltas) {
+      try { onDelta(d); } catch { /* a sink error must never abort the turn */ }
+    }
+  };
+
+  // Write prompt to a temp file (mode 0600) so the content never touches argv.
+  const tmpDir     = mkdtempSync(`${tmpdir()}/ocp-tui-`);
+  const promptFile = `${tmpDir}/prompt.txt`;
+  writeFileSync(promptFile, prompt, { mode: 0o600 });
+
  try {
-    // 1. Boot the interactive session inside tmux, rooted at the scratch cwd.
-    //    Capture the result: if tmux new-session fails (status !== 0) there is no
-    //    PTY, no interactive spawn — abort BEFORE the boot sleep rather than paste
-    //    into a non-existent session or issue a billing request without a verified
-    //    interactive context. The finally teardown is still harmless (kill-session
-    //    is a no-op when the session never existed).
-    const spawnResult = tmux(
-      ["new-session", "-d", "-s", tmuxName, "-x", "220", "-y", "50", "-c", cwd,
-       buildTuiCmd(claudeBin, model, sessionId, ehome, entrypointMode)],
-      { env },
-    );
-    if (!spawnResult || spawnResult.status !== 0) {
-      throw new Error("tui_spawn_failed: tmux session not created");
-    }
-
-    // 2. Wait until claude's input bar is actually ready (was: blind sleep(BOOT_MS)).
-    //    BOOT_MS is now the MAX readiness wait, not a fixed delay.
-    const ready = await pollUntil(() => tuiInputReady(tuiCapturePane(tmux, tmuxName)),
-      { timeoutMs: BOOT_MS, intervalMs: READY_POLL_MS });
-    if (!ready) {
-      // (readiness timed out; relying on paste-verify)
-      console.error("[tui] input_not_ready", tmuxName);
-    }
-
    // 3. Paste the prompt via a tmux PASTE BUFFER with bracketed paste (-p), NOT
    //    `send-keys -l`. send-keys of a large multi-line prompt is unreliable: the
    //    embedded newlines arrive as separate key events (effectively repeated Enter),
@@ -495,12 +746,37 @@ export async function runTuiTurn({
    // Submit (separate Enter key event).
    tmux(["send-keys", "-t", tmuxName, "Enter"]);

-    // 4. Block on the native transcript (resolved by session-id) until terminal.
-    //    Returns { text, entrypoint } from readTuiTranscript.
-    return await readTuiTranscript({ home: ehome, sessionId, wallclockMs });
+    // 5a. Streaming only: start polling the hook sink. Runs CONCURRENTLY with the
+    //     transcript wait below — the deltas are what make the answer visible while the
+    //     turn is still generating; the transcript is what makes it authoritative.
+    if (streaming) {
+      const loop = () => {
+        if (streamStopped) return;
+        drainDeltas();
+        pollTimer = setTimeout(loop, STREAM_POLL_MS);
+      };
+      pollTimer = setTimeout(loop, STREAM_POLL_MS);
+    }
+
+    // 5b. Block on the native transcript (resolved by THIS pane's session-id) until terminal.
+    //     Returns { text, entrypoint, truncated } from readTuiTranscript.
+    const result = await readTuiTranscript({ home: ehome, sessionId, wallclockMs, abortSignal });
+
+    // 5c. FINAL drain. The terminal marker can land between two poll ticks, so the last
+    //     delta(s) may still be unread — without this the tail would be missing from the
+    //     stream and every turn would need a transcript top-up.
+    streamStopped = true;
+    if (pollTimer) clearTimeout(pollTimer);
+    drainDeltas();
+    return result;
  } finally {
-    // 5. Teardown — always, even on throw.
+    // 6. Teardown — always, even on throw (including an abortSignal disconnect, which is
+    //    exactly why the pane cannot outlive a client that walked away). A pooled pane is
+    //    torn down here exactly like a cold-booted one: SINGLE-USE, never returned (pool.mjs).
+    streamStopped = true;
+    if (pollTimer) clearTimeout(pollTimer);
    try { tmux(["kill-session", "-t", tmuxName]); } catch { /* already gone */ }
    try { rmSync(tmpDir, { recursive: true, force: true }); } catch { /* best effort */ }
+    if (streamFile) { try { rmSync(streamFile, { force: true }); } catch { /* best effort */ } }
  }
 }
@@ -0,0 +1,273 @@
+// TUI-mode real SSE streaming — the `MessageDisplay` hook sink.
+//
+// WHAT THIS IS. `claude` fires a **MessageDisplay** hook per rendered block of the
+// assistant's reply, handing the hook the RAW MARKDOWN SOURCE of an incremental
+// `delta` on stdin. Registered via `--settings` on the ordinary interactive TUI spawn
+// (NO -p, NO --bare — the billing pool is untouched), it is the only byte-faithful
+// incremental source the interactive CLI exposes. Everything here consumes that hook
+// surface AS EMITTED — forwarding, not inventing.
+//
+// ALIGNMENT.md: **Class B**. We consume claude's own hook payload and re-emit it in the
+// OpenAI chat/completions streaming shapes OCP already speaks (ADR 0006). There is no
+// `cli.js` citation because no `cli.js` function is being mirrored: the TUI spawn is
+// OCP-owned surface (ADR 0007), and the hook payload is claude's own published contract.
+//
+// THE VERIFIED CONTRACT (docs/plans/2026-07-13-tui-latency/streaming-spike.md, and
+// independently reproduced on claude 2.1.207 / sonnet-4-6 / banner `· Claude Max`):
+//
+//   payload (stdin, one JSON object per fire):
+//     { hook_event_name:"MessageDisplay", session_id, transcript_path, prompt_id, cwd,
+//       turn_id, message_id, index, final, delta }
+//
+//   - deltas carry the raw markdown source (`## `, `**`, ```javascript all present)
+//   - concat(deltas of one message) === T, byte-exactly   (T = extractLatestAssistantText)
+//   - T.startsWith(concat(deltas[0..n])) at EVERY n       (prefix-stable)
+//   - block-level granularity (~5-7 fires per answer), NOT token-level
+//   - only `text` blocks fire it — thinking blocks are excluded (what OCP wants)
+//
+// ⚠️ THE HOOK IS SYNCHRONOUS. The hook's source sets `forceSyncExecution: true` —
+// `claude` BLOCKS on every fire. The hook script must therefore write and exit, doing
+// NO work inline. Measured cost of the script below: p50 7.2 ms / p90 14.7 ms per fire,
+// i.e. ~50 ms added blocking across a whole ~7-delta turn against a 6-10 s turn. That is
+// noise, so a plain append is the right sink — a FIFO would be faster on paper but a FIFO
+// blocks its writer until a reader attaches, which would hand `claude` a way to hang.
+//
+// WARM-POOL COMPATIBILITY (load-bearing — a warm pane pool is a separate in-flight PR).
+// The hook script and the settings file are BOTH STATIC: one copy per stream dir, written
+// once, never per-request. The per-turn destination is carried in the PANE'S OWN ENV as
+// `OCP_TUI_STREAM_FILE` (verified live: a hook inherits the pane's environment), and the
+// path is derived from the session-id — which for a pre-booted pane is fixed at BOOT.
+// Nothing about a request is baked into the settings file at spawn time, so a pane booted
+// before its request arrives streams exactly the same way.
+import { writeFileSync, mkdirSync, renameSync } from "node:fs";
+import { detectTuiUpstreamError } from "./transcript.mjs";
+
+// Default holdback before the first byte is released to the client. See TuiDeltaAssembler.
+export const DEFAULT_HOLDBACK_CHARS = 100;
+
+// The hook script. POSIX sh, no interpreter startup beyond /bin/sh, one fork (`cat`).
+//
+//   - `printf` is a shell BUILTIN in sh/dash/bash, so the newline costs no fork.
+//   - the `{ cat; printf '\n'; } >>` group opens the file ONCE and appends both writes
+//     through the same O_APPEND fd, so a payload and its terminator can never be split
+//     by another writer. (They never race anyway: one file per pane, and MessageDisplay
+//     is synchronous within a pane.)
+//   - a payload JSON can never contain a literal newline — JSON.stringify escapes them —
+//     so "one line == one payload" holds, and a torn write is always a trailing partial
+//     line, which parseDeltaChunk() leaves unconsumed until it completes.
+//   - NO OCP_TUI_STREAM_FILE (e.g. a pane booted with streaming off, or any other claude
+//     session that happens to load this settings file) => swallow stdin and exit 0. The
+//     hook must NEVER fail or block: claude is waiting on it.
+export const HOOK_SCRIPT = `#!/bin/sh
+# OCP TUI streaming sink — claude fires this per MessageDisplay block and BLOCKS on it.
+# Write and exit. Never do work here.
+[ -n "\$OCP_TUI_STREAM_FILE" ] || exec cat >/dev/null
+{ cat; printf '\\n'; } >> "\$OCP_TUI_STREAM_FILE"
+`;
+
+// The --settings payload registering the hook. Static: no per-request data.
+export function buildStreamSettings(hookScriptPath) {
+  return { hooks: { MessageDisplay: [{ hooks: [{ type: "command", command: hookScriptPath }] }] } };
+}
+
+export const hookScriptPath   = (streamDir) => `${streamDir}/md-hook.sh`;
+export const streamSettingsPath = (streamDir) => `${streamDir}/settings.json`;
+// One file per session-id. For a pre-booted (warm) pane the session-id is fixed at boot,
+// so this path is knowable at boot — which is what keeps the pool compatible.
+export const streamFilePath   = (streamDir, sessionId) => `${streamDir}/${sessionId}.jsonl`;
+
+// Atomic write: temp file + rename (same-directory, same-filesystem, so rename is atomic on
+// POSIX). A process killed mid-`writeFileSync` leaves the TEMP file half-written, never the
+// real path — `path` always names either the old complete content or the new complete
+// content, never a torn one. That matters specifically for md-hook.sh: it is SYNCHRONOUS
+// (claude blocks on every fire), so a truncated script would still pass `existsSync`, still
+// get exec'd, and fail/hang on every single MessageDisplay fire with no operator-visible
+// symptom short of streaming going silently dead (F7's streamZeroDeltaTurns is the backstop
+// for exactly that). Mirrors ensureTuiCwdTrusted's tmp+renameSync pattern in session.mjs.
+function writeFileAtomic(path, content, mode) {
+  const tmp = `${path}.${process.pid}.tmp`;
+  writeFileSync(tmp, content, { mode });
+  renameSync(tmp, path);
+}
+
+// Write the static hook script + settings file into `streamDir`. UNCONDITIONAL, not
+// write-if-missing: these files persist across OCP restarts at `streamDir`, so a host that
+// booted once under an older version and never had its stream dir cleared would otherwise be
+// silently stuck on a stale HOOK_SCRIPT / buildStreamSettings() forever — no future OCP
+// upgrade could ever reach it. Safe to call every boot: the content is static (no per-request
+// data), so a same-content rewrite is the overwhelmingly common case and costs two tiny
+// atomic writes, not a per-turn expense. Returns the settings path to hand to `claude
+// --settings`.
+export function prepareStreamHook(streamDir) {
+  mkdirSync(streamDir, { recursive: true });
+  const script = hookScriptPath(streamDir);
+  const settings = streamSettingsPath(streamDir);
+  writeFileAtomic(script, HOOK_SCRIPT, 0o700);
+  writeFileAtomic(settings, JSON.stringify(buildStreamSettings(script), null, 2), 0o600);
+  return settings;
+}
+
+// Parse newly-appended sink lines. `consumed` is the number of COMPLETE lines already
+// taken; only lines terminated by "\n" are complete, so a payload caught mid-write stays
+// unconsumed until its terminator lands. Returns the fresh MessageDisplay payloads plus
+// the new consumed count. Pure — the caller owns the cursor.
+export function parseDeltaChunk(text, consumed = 0) {
+  const lines = String(text ?? "").split("\n");
+  const complete = lines.slice(0, -1); // the tail after the last "\n" is a partial line
+  const deltas = [];
+  for (const line of complete.slice(consumed)) {
+    const t = line.trim();
+    if (!t) continue;
+    try {
+      const o = JSON.parse(t);
+      if (o && o.hook_event_name === "MessageDisplay" && typeof o.delta === "string") deltas.push(o);
+    } catch { /* not ours / not parseable — skip, never throw into the request path */ }
+  }
+  return { deltas, consumed: complete.length };
+}
+
+// ── The assembler: hook deltas → client bytes, with the honesty gates intact ──
+//
+// Two jobs, both load-bearing.
+//
+// 1. THE AUTH-BANNER HOLDBACK (C-1 / issue #133 must survive streaming).
+//    The interactive CLI renders an auth failure as ordinary assistant TEXT — so an
+//    expired-credential turn fires MessageDisplay with the BANNER as its delta, and a
+//    naive forwarder would stream "Please run /login · API Error: 401 …" to the client as
+//    a normal answer, exactly the silent-error case C-1 exists to prevent.
+//    detectTuiUpstreamError() classifies a WHOLE message, so it cannot be run per-delta.
+//    Instead we HOLD BACK the first `holdbackChars` characters. The default detector only
+//    ever fires on a message of <= 100 chars (TUI_ERR_MAX_LEN — real banners are 69 and 73),
+//    so once the TRIMMED accumulation EXCEEDS 100 chars the final text cannot be a banner by
+//    that detector's own length rule, and releasing is safe. An answer that never exceeds the
+//    holdback is simply delivered whole at terminal — i.e. exactly today's buffered
+//    behaviour, gates and all.
+//    THE GUARANTEE HAS TWO HALVES, both required — neither alone is sufficient:
+//      (i)  Nothing is emitted for a message until its trimmed accumulation exceeds the
+//           detector's max banner length. This is what keeps the FIRST message of a turn
+//           safe: a banner-length message can never clear the holdback.
+//      (ii) Once a message boundary follows an emit (`restartedAfterEmit`), push() stops
+//           emitting ENTIRELY for the rest of the turn — a SECOND message (e.g. an
+//           auth-failure banner rendered mid-turn, after tool-using prose already streamed)
+//           gets zero bytes forwarded, not just a fresh holdback of its own. finalize() then
+//           refuses the whole turn (SSE error frame, no cache) precisely because the first
+//           message's bytes are unretractable and unverifiable against T. Without this half,
+//           (i) alone only protects the FIRST message per turn — see F1.
+//    ⚠️ Soundness is w.r.t. the DEFAULT detector. An operator who REPLACES it via
+//    CLAUDE_TUI_ERROR_PATTERNS with a pattern that can match a longer message must raise
+//    OCP_TUI_STREAM_HOLDBACK past their longest banner; server.mjs warns at boot. That is the
+//    one case (i) does not cover — (ii) still applies regardless. Even past both, the
+//    terminal gate still refuses to cache a banner and still ends the stream on an SSE error
+//    frame rather than finish_reason:"stop" — the holdback is the first of two layers, not
+//    the only one.
+//
+// 2. MESSAGE SCOPING (keeps `concat === T` the RIGHT assertion).
+//    The transcript's T is extractLatestAssistantText() — the LAST text-bearing assistant
+//    entry, not every assistant entry. A tool-using turn therefore has TWO messages
+//    (prose → tool_use → answer) and T is only the second. So the assembler scopes to the
+//    CURRENT message_id: when a new message_id appears and NOTHING has been emitted yet,
+//    the held text is DISCARDED — the transcript is about to discard it too, so this keeps
+//    us byte-identical to the buffered path instead of streaming prose the buffered path
+//    would have dropped. When a new message_id appears AFTER we have already emitted, the
+//    bytes are gone and cannot be retracted: finalize() then reports !ok and the caller
+//    fails the turn loudly (SSE error frame, no cache, counted on /health). Fail-loud is
+//    the correct posture — a proxy that silently serves text the transcript disagrees with
+//    is the exact class of bug ALIGNMENT.md exists to prevent.
+// Sentinel for "no message seen yet". Deliberately not null/undefined — see the constructor.
+const NO_MESSAGE_YET = Symbol("no-message-yet");
+
+export class TuiDeltaAssembler {
+  constructor({ holdbackChars = DEFAULT_HOLDBACK_CHARS, detectError = detectTuiUpstreamError } = {}) {
+    this.holdbackChars = holdbackChars;
+    this.detectError = detectError;
+    this.emitted = "";   // bytes ALREADY written to the client — unretractable
+    this.pending = "";   // held back, not yet written
+    this.released = false;
+    // NOT null: a payload may legitimately carry message_id === null, and if the sentinel were
+    // also null the FIRST such payload would compare equal to it, register no boundary, and
+    // leave `messages` at 0 — which used to disarm the restartedAfterEmit guard below entirely.
+    // A unique object is === to nothing a JSON payload can produce, so the first fire ALWAYS
+    // registers as message 1, whatever its message_id is (or isn't).
+    this.messageId = NO_MESSAGE_YET;
+    this.deltas = 0;     // hook fires seen
+    this.messages = 0;   // distinct message_ids seen
+    this.restartedAfterEmit = false;
+  }
+
+  // All hook bytes for the CURRENT message (emitted + still held).
+  get full() { return this.emitted + this.pending; }
+
+  // Feed one MessageDisplay payload. Returns the text to emit NOW, or null (held back).
+  push(payload) {
+    const delta = payload && typeof payload.delta === "string" ? payload.delta : "";
+    const mid = payload ? payload.message_id : null;
+    if (mid !== this.messageId) {
+      this.messageId = mid;
+      this.messages++;
+      if (this.emitted === "") {
+        this.pending = "";        // safe: the transcript will drop this message too
+      } else {
+        // A boundary while bytes are ALREADY out is unrecoverable, full stop — the count of
+        // messages seen so far is irrelevant. The old `else if (this.messages > 1)` guard was
+        // the sole reason a null-message_id first payload could disarm F1: it left `messages`
+        // at 0, so the real boundary evaluated 1 > 1 === false and never armed. The invariant
+        // is "a boundary occurred while emitted !== ''", and that is exactly what this says.
+        this.restartedAfterEmit = true; // unrecoverable — finalize() will refuse the turn
+      }
+    }
+    this.deltas++;
+    // F1: once a message boundary has followed an emit, the turn is ALREADY unrecoverable —
+    // finalize() will refuse it (see restartedAfterEmit above). `this.released` stays true
+    // from the FIRST message's release and, uncorrected, lets every later message's deltas
+    // stream straight through unfiltered — exactly the auth-banner-mid-turn leak this class
+    // exists to prevent. Stop emitting HERE, permanently, for the rest of the turn: there is
+    // nothing left to gain from continuing to forward bytes for a turn that will be refused,
+    // and every byte forwarded now is one more the client cannot be told to un-see.
+    if (this.restartedAfterEmit) return null;
+    if (!delta) return null;
+
+    if (this.released) {
+      this.emitted += delta;
+      return delta;
+    }
+    this.pending += delta;
+    // Release only once the TRIMMED accumulation is past the banner detector's reach.
+    // detectTuiUpstreamError() trims before measuring length (TUI_ERR_MAX_LEN is a trimmed-
+    // length bound), so gating release on the UNTRIMMED pending.length let a run of >
+    // holdbackChars whitespace trim down to "" — detectError("") sees nothing to classify,
+    // returns null, and release fires with the holdback never having actually screened
+    // anything. Trimming here keeps both sides of the check talking about the same string.
+    if (this.pending.trim().length > this.holdbackChars && this.detectError(this.pending) == null) {
+      const out = this.pending;
+      this.pending = "";
+      this.released = true;
+      this.emitted += out;
+      return out;
+    }
+    return null;
+  }
+
+  // Reconcile against the AUTHORITATIVE transcript text T. Call only AFTER the truncation
+  // and auth-banner gates have passed. Returns:
+  //   { ok:true,  tail, exact }  — tail is the remaining text to emit (may be ""). `exact`
+  //                               is concat(deltas) === T; when false we still serve exactly
+  //                               T, having topped up from the transcript, and the caller
+  //                               counts a topUp.
+  //   { ok:false, ... }          — what we already emitted is NOT a prefix of T. The client
+  //                               holds bytes the transcript disagrees with; the caller must
+  //                               NOT cache and must end the stream on an SSE error frame.
+  finalize(T) {
+    const text = typeof T === "string" ? T : "";
+    const full = this.full;
+    if (!text.startsWith(this.emitted)) {
+      return { ok: false, tail: null, exact: false, emitted: this.emitted.length, transcript: text.length };
+    }
+    return {
+      ok: true,
+      tail: text.slice(this.emitted.length),
+      exact: full === text,
+      emitted: this.emitted.length,
+      transcript: text.length,
+    };
+  }
+}
@@ -73,6 +73,17 @@ export function isTerminalLine(obj) {
 // transcript holding one logical exchange). If a future warm-pool ever reuses a
 // session WITHOUT a fresh session-id / clear, earlier-turn text could leak — that
 // author must add user-line scoping here. See spec §7.2.
+//
+// STATUS (warm pool, lib/tui/pool.mjs — the "future warm-pool" this note anticipated):
+// the pool does NOT reuse sessions, so the precondition above still holds and no
+// user-line scoping was added. Each pooled pane is booted with its OWN fresh
+// randomUUID() --session-id (bootTuiPane) and is SINGLE-USE: it serves exactly one turn
+// and is then killed and replaced. One session still means one logical exchange, so the
+// last assistant entry is still that request's answer.
+// The warning therefore stands UNCHANGED for anyone who later wants a pane to serve a
+// SECOND turn (or to reset one with /clear and reuse it): that is a leak, and it needs
+// user-line scoping HERE before it can be safe. Do not relax pool.mjs's single-use rule
+// without doing that work first.
 export function extractLatestAssistantText(events) {
  let text = "";
  for (const ev of events) {
@@ -256,11 +267,21 @@ export function detectTuiUpstreamError(text, patternsRaw = process.env.CLAUDE_TU
 // Resolution: pass an explicit `transcriptPath` (used by unit tests), OR pass
 // `home` + `sessionId` to resolve by glob each poll (production) — the transcript
 // file does not exist until the turn starts, so resolution happens inside the loop.
-export async function readTuiTranscript({ transcriptPath: p, home, sessionId, wallclockMs = 120000, pollMs = 250 }) {
+// `abortSignal` (optional): when it fires, stop waiting and throw TuiAbortError. The one
+// caller that passes it is the STREAMING TUI path, which ties it to the client's socket:
+// a client that disconnects mid-turn should not leave the pane running (and the caller's
+// concurrency slot held) until the turn or the 120s cap ends. runTuiTurn's finally does the
+// teardown. Omitted => the loop is byte-for-byte the pre-streaming loop.
+export async function readTuiTranscript({ transcriptPath: p, home, sessionId, wallclockMs = 120000, pollMs = 250, abortSignal = null }) {
  const deadline = Date.now() + wallclockMs;
  let lastText = "";
  let lastEntrypoint = null;
  while (Date.now() < deadline) {
+    if (abortSignal && abortSignal.aborted) {
+      const err = new Error("tui_aborted: client disconnected before the turn completed");
+      err.name = "TuiAbortError";
+      throw err;
+    }
    const resolved = p || findTranscriptPath(home, sessionId);
    if (resolved && existsSync(resolved)) {
      const events = parseTranscriptLines(readFileSync(resolved, "utf8"));
@@ -22,6 +22,8 @@
 *   CLAUDE_MAX_CONCURRENT        — max concurrent claude processes, -p/stream-json path (default: 8)
 *   CLAUDE_MAX_QUEUE             — max requests waiting for a -p slot before HTTP 429 (default: 16)
 *   OCP_TUI_MAX_CONCURRENT       — max concurrent interactive TUI turns, TUI-mode path (default: 2)
+ *   OCP_TUI_POOL_SIZE            — pre-booted warm `claude` panes held for TUI-mode (default: 0 = off;
+ *                                  max 4). Each is a live idle process; cuts ~3-4s per request.
 *   OCP_SPAWN_REAL_HOME          — "1" forces the -p spawn to use the real HOME (disables the
 *                                  latency spawn-home isolation; default: isolated when a token exists)
 *   CLAUDE_BREAKER_THRESHOLD     — failures in window before circuit opens (default: 6)
@@ -32,7 +34,7 @@
 *   CLAUDE_HEARTBEAT_INTERVAL    — SSE heartbeat interval in ms on streaming path (default: 0 = disabled)
 */
 import { createServer } from "node:http";
-import { spawn, execFileSync } from "node:child_process";
+import { spawn, execFileSync, spawnSync } from "node:child_process";
 import { randomUUID, timingSafeEqual } from "node:crypto";
 import { readFileSync, readdirSync, accessSync, existsSync, constants, chmodSync, statSync, mkdirSync, writeFileSync, rmSync } from "node:fs";
 import { fileURLToPath } from "node:url";
@@ -41,9 +43,11 @@ import { homedir } from "node:os";
 import { validateKey, recordUsage, getUsageByKey, getUsageTimeline, getRecentUsage, createKey, listKeys, revokeKey, closeDb, checkQuota, updateKeyQuota, getKeyQuota, findKey, cacheHash, getCachedResponse, setCachedResponse, clearCache, getCacheStats, hasCacheControl, singleflight, getInflightStats } from "./keys.mjs";
 import { DEFAULT_PORT } from "./lib/constants.mjs";
 import { isLoopbackBind } from "./lib/net.mjs";
-import { runTuiTurn, reapStaleTuiSessions, resolveTuiHome } from "./lib/tui/session.mjs";
+import { runTuiTurn, reapStaleTuiSessions, resolveTuiHome, bootTuiPane, tuiPaneHealthy, poolPaneName, POOL_BOOT_MS } from "./lib/tui/session.mjs";
 import { detectTuiUpstreamError } from "./lib/tui/transcript.mjs";
 import { TuiSemaphore, SemaphoreAbortError, recordTuiEntrypoint, buildTuiHealthBlock } from "./lib/tui/semaphore.mjs";
+import { TuiPanePool, resolvePoolSize, POOL_MAX_SIZE } from "./lib/tui/pool.mjs";
+import { TuiDeltaAssembler, DEFAULT_HOLDBACK_CHARS } from "./lib/tui/stream.mjs";
 import { createSerialMutex, createTtlCache, isTokenExpiring, orderLabelsLastGoodFirst } from "./lib/spawn-auth.mjs";

 const __dirname = dirname(fileURLToPath(import.meta.url));
@@ -349,8 +353,98 @@ const tuiSemaphore = new TuiSemaphore(TUI_MAX_CONCURRENT);
 const tuiStats = {
  lastEntrypoint: null,      // last observed cc_entrypoint from the transcript ("cli" | "sdk-cli" | null)
  entrypointMismatches: 0,   // count of cli-expected-but-got-other turns
+  streamTurns: 0,            // streamed TUI turns ATTEMPTED (counted before the honesty gates — F6)
+  streamDeltas: 0,           // MessageDisplay hook fires OBSERVED (forwarded + held-back — F6)
+  streamTopUps: 0,           // turns where the delta stream != T but was a safe PREFIX of it
+  streamDivergences: 0,      // turns REFUSED: emitted bytes were not a prefix of T
+  streamZeroDeltaTurns: 0,   // streamed turns where the hook fired ZERO times (F7 — the hook is
+                             // dead, not just one fire dropped; distinct from streamTopUps)
 };

+// ── TUI real streaming (backlog #2) — opt-in; default OFF ────────────────
+// When ON *and* TUI_MODE is on *and* the client asked for stream:true, the turn is emitted
+// as real SSE delta.content chunks as claude renders them, sourced from claude's own
+// MessageDisplay hook (lib/tui/stream.mjs). When OFF, the buffered
+// callClaudeTui → streamStringAsSSE path below is byte-for-byte unchanged — the spawn does
+// not even get --settings. Opt-in is deliberate: the buffered path is stable production.
+//
+// Honest expectation (docs/plans/2026-07-13-tui-latency/streaming-spike.md): this moves the
+// FIRST byte, not the last. A consumer that must parse a complete reply gains nothing; a
+// progressively-rendering chat UI gains the ~4s between first delta and last. It does not
+// move the ~6s TTFT floor of TUI mode.
+const TUI_STREAM = process.env.OCP_TUI_STREAM === "1";
+const TUI_STREAM_DIR = process.env.OCP_TUI_STREAM_DIR || `${process.env.HOME}/.ocp-tui/stream`;
+// First-bytes holdback — the auth-banner gate's (C-1) survival mechanism under streaming.
+// See TuiDeltaAssembler: nothing is emitted for a message until its TRIMMED accumulation
+// exceeds this, which puts it out of the default banner detector's <=100-char reach — the
+// FIRST of the two halves of the guarantee (see the assembler's class comment for the second:
+// no further emission at all once a message boundary follows an emit). Only raise it.
+const TUI_STREAM_HOLDBACK = parseInt(process.env.OCP_TUI_STREAM_HOLDBACK || String(DEFAULT_HOLDBACK_CHARS), 10);
+if (TUI_MODE && TUI_STREAM && process.env.CLAUDE_TUI_ERROR_PATTERNS != null && TUI_STREAM_HOLDBACK <= DEFAULT_HOLDBACK_CHARS) {
+  // The holdback's FIRST-MESSAGE half (see TuiDeltaAssembler) is sound for the DEFAULT
+  // auth-banner detector (which cannot match a message longer than 100 chars). An
+  // operator-supplied pattern set has no such bound, so a banner longer than the holdback
+  // could reach the client before the terminal gate rejects the turn. (The second half — no
+  // further emission once a message boundary follows an emit — holds regardless of the
+  // detector; this warning is only about the first-message case.)
+  console.error(
+    `[tui] WARNING: OCP_TUI_STREAM=1 with a custom CLAUDE_TUI_ERROR_PATTERNS and holdback=${TUI_STREAM_HOLDBACK}.\n` +
+    "  The streaming holdback's first-message coverage is sound only against the DEFAULT banner\n" +
+    "  detector (<=100 chars). Raise OCP_TUI_STREAM_HOLDBACK above your longest custom banner, or\n" +
+    "  the first chars of one could be streamed before the end-of-turn gate refuses the turn."
+  );
+}
+
+// ── Warm pane pool (docs/plans/2026-07-13-tui-latency #3) — opt-in; default OFF ─────────
+// OCP_TUI_POOL_SIZE=0 (default) => tuiPool is null => runTuiTurn's cold-boot path is
+// byte-for-byte unchanged. Set it to N (clamped to POOL_MAX_SIZE) to keep N pre-booted
+// `claude` panes warm, each SINGLE-USE (see lib/tui/pool.mjs for why single-use is the
+// load-bearing rule, and lib/tui/session.mjs for the POOL/REAPER INVARIANT).
+//
+// Default-off is deliberate on a stable production path: a warm pane is a LIVE idle
+// `claude` process held whether or not a request ever arrives, so the operator must opt
+// in to that standing cost. Measured saving when on (this host, Sonnet 4.6, --effort low):
+// end-to-end p50 10.17 s (n=6, pool off) -> 6.00 s (n=12 warm hits), i.e. -41%.
+// cli.js does NOT perform this operation (Class B, OCP-owned TUI spawn) — see ADR 0007.
+const TUI_POOL_SIZE = TUI_MODE ? resolvePoolSize(process.env.OCP_TUI_POOL_SIZE) : 0;
+const tuiPool = TUI_POOL_SIZE > 0
+  ? new TuiPanePool({
+      size: TUI_POOL_SIZE,
+      // The POOL mints the pane's identity, not bootTuiPane: the tmux session exists the
+      // instant the boot starts, so the pool must be able to name (hence spare, hence kill)
+      // it before then. Name is derived from the session-id, so `tmux ls` correlates to the
+      // transcript file <HOME>/.claude/projects/*/<sessionId>.jsonl.
+      mintPane: () => {
+        const sessionId = randomUUID();
+        return { sessionId, name: poolPaneName(PORT, sessionId) };
+      },
+      bootPane: (model, ident) => bootTuiPane({
+        model,
+        claudeBin: CLAUDE,
+        home: TUI_HOME,
+        realHome: process.env.HOME,
+        cwd: TUI_CWD,
+        port: PORT,
+        entrypointMode: TUI_ENTRYPOINT,
+        sessionId: ident.sessionId,
+        name: ident.name,
+        requireReady: true,     // a pane that never reached its input bar must not be enlisted
+        bootMs: POOL_BOOT_MS,   // background pre-boot — no client is blocked, so be patient
+        // Warm panes must carry the MessageDisplay hook too, or every pool HIT would
+        // silently fall back to buffered while every MISS streamed — the two paths have to
+        // spawn identically (F4). Gated on TUI_STREAM, the deployment-wide switch — NOT on any
+        // particular request's stream:true/false, which does not exist yet at pre-boot time.
+        // The runTuiTurn cold-boot call site (callClaudeTui, below) mirrors this exact gate for
+        // the same reason. bootTuiPane derives the sink from the pane's own session-id, which is
+        // minted above, so nothing request-specific is baked in at pre-boot time.
+        streamDir: TUI_STREAM ? TUI_STREAM_DIR : null,
+      }),
+      killPane: (name) => { try { spawnSync(process.env.OCP_TUI_TMUX_BIN || "tmux", ["kill-session", "-t", name]); } catch { /* already gone */ } },
+      paneHealthy: (name) => tuiPaneHealthy((args) => spawnSync(process.env.OCP_TUI_TMUX_BIN || "tmux", args, { encoding: "utf8" }), name),
+      log: (level, event, data) => logEvent(level, event, data),
+    })
+  : null;
+
 // ── FIX ③ (latency): default-path (-p / stream-json) spawn-home isolation ──────────────
 // PROBLEM (measured, not theoretical): OCP's default spawn inherits the operator's real HOME
 // (loading the global ~/.claude — plugins, skills, hooks) and runs with cwd=~/ocp (loading the
@@ -770,19 +864,45 @@ const cacheCleanupInterval = setInterval(() => {
 // mechanism and the 15-min cadence makes the window negligible).
 // Gated on TUI_MODE — zero effect (no kill-server, no list-sessions) when TUI is off.
 // cli.js does NOT perform this operation (Class B, OCP-owned TUI spawn) — see ADR 0007.
+//
+// WARM POOL INTERACTION (the crux — see the POOL/REAPER INVARIANT in lib/tui/session.mjs).
+// A warm pooled pane is one of OUR OWN ocp-tui-<port>-* sessions that is alive and idle BY
+// DESIGN, and this sweep fires precisely when the instance is idle — i.e. exactly when the
+// pool is full. Two things are therefore required, and both are done here:
+//   (a) DRAIN the pool BEFORE the sweep. Zombie reaping is possible ONLY via kill-server,
+//       and a live pooled pane suppresses kill-server (it is a live child of the tmux
+//       server). A permanently-full pool would otherwise permanently disable the very
+//       thing this tick exists to do. Draining costs one pane re-boot per tick (~1.2 s of
+//       background work every 15 min) and is invisible to callers: a request landing in the
+//       drain→refill gap simply MISSES the pool and takes today's cold path.
+//   (b) Pass the pool's live registry as `spare` anyway. After (a) it is empty, so this is
+//       belt-and-braces — it makes it impossible for THIS call site (or a future one) to
+//       kill a live pooled pane even if the drain were ever removed or reordered.
+// RESIDUAL (unchanged in kind from the pre-pool code, and explicitly accepted there): a
+// request arriving in the narrow window between the idle-check and kill-server has its pane
+// torn down and fails cleanly via runTuiTurn's honesty gates. The drain widens that window
+// by the cost of N kill-session calls (single-digit ms), not materially.
 const TUI_REAP_INTERVAL_MS = 15 * 60 * 1000;
 const tuiReapInterval = TUI_MODE ? setInterval(() => {
  if (tuiSemaphore.inflight > 0 || tuiSemaphore.queued > 0) return; // a turn is live — defer
  try {
+    const drained = tuiPool ? tuiPool.drain() : 0;
    // F7 fix: scope to THIS instance's own port; a sibling ocp-tui-<otherPort>-* session
    // (a second OCP instance on the same host) is treated as foreign, same as olp-tui-*.
    // includeLegacy is NOT set here — see reapStaleTuiSessions' comment: the periodic sweep
    // conservatively treats any lingering bare-prefix legacy session as foreign so it can
    // never trigger kill-server on a steady-state tick; only the one-time boot reap below
    // claims legacy-shaped zombies.
-    const n = reapStaleTuiSessions({ port: PORT });
-    if (n) logEvent("info", "tui_reaped_stale_sessions", { count: n, trigger: "periodic" });
+    const n = reapStaleTuiSessions({ port: PORT, spare: tuiPool ? tuiPool.liveNames() : null });
+    if (n || drained) {
+      logEvent("info", "tui_reaped_stale_sessions", { count: n, poolDrained: drained, trigger: "periodic" });
+    }
  } catch (e) { logEvent("error", "tui_periodic_reap_failed", { error: e.message }); }
+  finally {
+    // Refill in the background regardless of how the sweep went — a throw mid-sweep must not
+    // leave the pool permanently paused (it would silently degrade to the cold path forever).
+    if (tuiPool) { try { tuiPool.resume(); } catch { /* best effort */ } }
+  }
 }, TUI_REAP_INTERVAL_MS) : null;
 if (tuiReapInterval && typeof tuiReapInterval.unref === "function") tuiReapInterval.unref();

@@ -1283,7 +1403,14 @@ async function callClaude(model, messages, conversationId, keyName, res) {
 // Authority: claude CLI v2.1.158 interactive mode (cc_entrypoint=cli).
 // SECURITY: A-path single-user ONLY — home is NOT isolation (see ADR 0007).
 // `res` (optional, F2) is the client's http.ServerResponse — see closeSignalFor.
-async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
+//
+// `streamCtx` (optional, OCP_TUI_STREAM): { emit(text), signal } — when present the turn is
+// ALSO streamed live via claude's MessageDisplay hook. The contract is unchanged: this still
+// returns the TRANSCRIPT's text (T), the honesty gates still run on T before anything is
+// committed, and the cache still stores T — never the concatenated deltas. streamCtx.emit is
+// the SSE sink; streamCtx.signal is the client's disconnect signal, which tears the pane down
+// mid-turn instead of holding the semaphore slot for a dead socket.
+async function callClaudeTui(model, messages, _conversationId, _keyName, res, streamCtx = null) {
  const cliModel = MODEL_MAP[model] || model;
  const prompt = messagesToPrompt(messages); // includes system as [System] inline
  recordModelRequest(cliModel, prompt.length);
@@ -1311,6 +1438,23 @@ async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
  // release() runs in a finally so any throw from runTuiTurn (tmux spawn failure,
  // paste-not-landed) OR from the honesty gates below (truncation / error banner) can NEVER
  // leak a slot. tuiSemaphore.inflight feeds /health.
+  // Streaming assembler (null when OCP_TUI_STREAM is off — then runTuiTurn gets no onDelta,
+  // spawns no hook, and behaves byte-for-byte as before). It owns the auth-banner holdback
+  // and the message scoping; see lib/tui/stream.mjs.
+  const assembler = streamCtx ? new TuiDeltaAssembler({ holdbackChars: TUI_STREAM_HOLDBACK }) : null;
+  // F6: counted here — the moment a streamed turn is ATTEMPTED — not after the honesty gates
+  // below. A turn refused by the truncation or auth-banner gate is exactly the turn an operator
+  // most wants visible in streamTurns; counting only turns that reached the gates made
+  // streamDivergences/streamTurns silently exclude its own worst cases from the denominator.
+  if (assembler) tuiStats.streamTurns++;
+  const onDelta = assembler
+    ? (payload) => {
+        const out = assembler.push(payload);
+        tuiStats.streamDeltas++; // every hook fire OBSERVED, not just forwarded ones — see the
+                                  // /health field doc in lib/tui/semaphore.mjs (F6)
+        if (out) streamCtx.emit(out); // released past the holdback — safe to show the client
+      }
+    : null;
  try {
    const { text, entrypoint, truncated } = await runTuiTurn({
      prompt,
@@ -1323,6 +1467,27 @@ async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
                  // different port never collides with this instance's reap/kill-server logic.
      wallclockMs: TUI_WALLCLOCK_MS,
      entrypointMode: TUI_ENTRYPOINT,
+      // Warm pane pool (null unless OCP_TUI_POOL_SIZE > 0 → today's cold path exactly).
+      // A pooled pane is single-use: runTuiTurn kills it in its finally like any other.
+      pool: tuiPool,
+      // Only observe when the pool is ON — with it off (the default) no new log line is
+      // emitted, so the disabled path stays byte-for-byte today's, logs included.
+      onPane: tuiPool
+        ? ({ warm }) => logEvent("info", warm ? "tui_pool_hit" : "tui_pool_miss",
+            { model: cliModel, warmRemaining: tuiPool.warm })
+        : null,
+      onDelta,
+      // Gated on TUI_STREAM (the deployment-wide switch), NOT on `assembler` (this REQUEST's
+      // stream:true/false) — F4 fix. The pool's bootPane closure above installs the hook on
+      // every warm pane whenever TUI_STREAM is on, regardless of what any given future request
+      // asks for (a pre-booted pane cannot know that yet); the cold path must match, or a
+      // stream:false request gets --settings on a pool HIT and not on a pool MISS — two
+      // different spawn argvs for the identical request, which this project's alignment/billing
+      // posture cannot tolerate. Whether the hook's OUTPUT is actually consumed for THIS turn is
+      // decided downstream by `onDelta` (null when assembler is null), so a non-streaming
+      // request still never polls or emits — it just spawns identically either way.
+      streamDir: TUI_STREAM ? TUI_STREAM_DIR : null,
+      abortSignal: streamCtx ? streamCtx.signal : null,
    });
    // ── Honesty gates (issue #133) ─ run BEFORE recordModelSuccess / cache write-back.
    // A throw here propagates to the catch below (recordModelError + reject), so the
@@ -1347,6 +1512,59 @@ async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
      throw new Error("tui_upstream_error: claude CLI returned an in-session error banner instead of an answer");
    }

+    // ── Streaming safety net — the transcript is the authority, the deltas are the mirror.
+    // Runs AFTER the two gates above (so a truncated turn or an auth banner is never
+    // reconciled, let alone flushed) and BEFORE recordModelSuccess / the caller's cache
+    // write. Three outcomes:
+    //   exact     — concat(deltas) === T. The invariant held; emit whatever is still held
+    //               back (a short answer never passes the holdback, so this is its whole text).
+    //   top-up    — what we emitted is a strict PREFIX of T but the deltas did not add up to
+    //               it (a dropped/late fire). We serve exactly T by emitting the missing tail;
+    //               the client still gets the right answer. Counted, and visible on /health.
+    //   divergence— we already emitted bytes that are NOT a prefix of T. The client is holding
+    //               text the transcript disagrees with and it cannot be retracted. REFUSE the
+    //               turn: throw → SSE error frame, no cache, no success. Serving on would be
+    //               exactly the "silently serve wrong text" failure this gate exists to stop.
+    //               (Known trigger: a tool-using turn whose pre-tool prose exceeded the
+    //               holdback — the transcript keeps only the LAST assistant message, so the
+    //               prose we streamed is text T does not contain.)
+    if (assembler) {
+      // F7: a total hook failure (a claude version bump stops honoring --settings, or a
+      // truncated md-hook.sh per F3) produces zero fires for every turn, finalize() still
+      // reports ok:true/exact:false (the transcript alone carries the whole answer), and the
+      // turn succeeds NORMALLY — degrading to buffered with no error, no divergence, nothing
+      // but streamTopUps climbing (which the comment above calls "benign"). That is
+      // indistinguishable from one late fire dropped unless it is counted separately.
+      if (assembler.deltas === 0) {
+        tuiStats.streamZeroDeltaTurns++;
+        logEvent("warn", "tui_stream_zero_deltas", { model: cliModel });
+      }
+      const rec = assembler.finalize(text);
+      if (!rec.ok) {
+        tuiStats.streamDivergences++;
+        logEvent("error", "tui_stream_divergence", {
+          model: cliModel,
+          // The dominant cause in practice: a TOOL-USING turn whose pre-tool prose exceeded the
+          // holdback and was already streamed. The transcript keeps only the LAST assistant
+          // message, so that prose is text T does not contain. Remedy for such a deployment:
+          // raise OCP_TUI_STREAM_HOLDBACK above the model's typical narration length (later first
+          // chunk, but the prose stays held back and is then correctly discarded), or leave
+          // OCP_TUI_STREAM off. See README + ADR 0007 (2026-07-13 amendment).
+          reason: assembler.restartedAfterEmit ? "multi_message_after_emit (tool-use turn?)" : "delta_transcript_mismatch",
+          emittedChars: rec.emitted, transcriptChars: rec.transcript,
+          deltas: assembler.deltas, messages: assembler.messages,
+        });
+        throw new Error("tui_stream_divergence: streamed text is not a prefix of the transcript; refusing to serve it");
+      }
+      if (!rec.exact) {
+        tuiStats.streamTopUps++;
+        logEvent("warn", "tui_stream_topup", {
+          model: cliModel, emittedChars: rec.emitted, transcriptChars: rec.transcript, deltas: assembler.deltas,
+        });
+      }
+      if (rec.tail) streamCtx.emit(rec.tail);
+    }
+
    recordModelSuccess(cliModel, 0); // elapsed not measurable here; wallclock at reader level
    // Assert the subscription-pool classification. TUI exists to keep cc_entrypoint=cli
    // (subscription pool); a silent degrade to sdk-cli (metered Agent SDK pool) would still
@@ -1361,6 +1579,14 @@ async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
    }
    return text;
  } catch (err) {
+    // A mid-turn client disconnect (streaming path only — abortSignal) is NOT an upstream
+    // failure: runTuiTurn's finally already tore the pane down, and this finally releases the
+    // slot. Mirror the queued-disconnect handling above (info, no recordModelError, no
+    // response) rather than booking a phantom model error against the socket going away.
+    if (err && err.name === "TuiAbortError") {
+      logEvent("info", "tui_turn_aborted", { reason: "client_disconnected", model: cliModel });
+      throw new RequestDisconnectedError("client disconnected mid-turn; TUI pane torn down");
+    }
    recordModelError(cliModel, false);
    throw err;
  } finally {
@@ -1368,6 +1594,102 @@ async function callClaudeTui(model, messages, _conversationId, _keyName, res) {
  }
 }

+// ── TUI-mode REAL streaming (OCP_TUI_STREAM=1) ──────────────────────────
+// The stream:true + TUI_MODE + OCP_TUI_STREAM=1 path. Emits the turn as it is generated,
+// from claude's own MessageDisplay hook, instead of buffering it and replaying it with
+// streamStringAsSSE.
+//
+// WIRE SHAPES: every frame below is COPIED from callClaudeStreaming (the -p path) — the role
+// chunk, the content-delta chunk, the stop chunk, `[DONE]`, and the post-header
+// {error:{message,type}} frame. No new fields, no new shapes. (ALIGNMENT.md Rule 2 / Class B:
+// the authority for the wire format is the OpenAI chat/completions streaming spec, adopted by
+// ADR 0006; the authority for the TUI spawn is ADR 0007. No cli.js citation applies — see the
+// commit body.)
+//
+// HEADERS ARE SENT EAGERLY, exactly as the -p path does, so the existing heartbeat
+// (CLAUDE_HEARTBEAT_INTERVAL) covers the ~6s of silence before the first delta. The cost is
+// the same one the -p path already pays: after the headers are out, an upstream failure can
+// no longer be a JSON 500, so it is surfaced as the SSE error frame instead (issue #110).
+async function callClaudeTuiStreaming(model, messages, conversationId, res, authInfo = {}) {
+  const id = `chatcmpl-${randomUUID()}`;
+  const created = Math.floor(Date.now() / 1000);
+  const t0 = Date.now();
+  const promptChars = messages.reduce((a, m) => a + contentToText(m.content).length, 0);
+  let headersSent = false;
+
+  function ensureHeaders() {
+    if (res.writableEnded || res.destroyed) return false;
+    if (headersSent) return true;
+    headersSent = true;
+    res.writeHead(200, {
+      "Content-Type": "text/event-stream",
+      "Cache-Control": "no-cache",
+      "Connection": "keep-alive",
+      "X-Accel-Buffering": "no",
+    });
+    sendSSE(res, {
+      id, object: "chat.completion.chunk", created, model,
+      choices: [{ index: 0, delta: { role: "assistant" }, finish_reason: null }],
+    });
+    return true;
+  }
+
+  ensureHeaders();
+  const hb = startHeartbeat(res, HEARTBEAT_INTERVAL, conversationId);
+  // Held for the WHOLE turn (not just the queue wait): a disconnect must abort the transcript
+  // wait so runTuiTurn tears the pane down and callClaudeTui's finally frees the slot.
+  const { signal, detach } = closeSignalFor(res);
+
+  const streamCtx = {
+    signal,
+    emit(text) {
+      if (!text) return;
+      if (!ensureHeaders()) return; // client vanished — drop the write, the turn still unwinds
+      sendSSE(res, {
+        id, object: "chat.completion.chunk", created, model,
+        choices: [{ index: 0, delta: { content: text }, finish_reason: null }],
+      }, hb);
+    },
+  };
+
+  try {
+    // callClaudeTui returns the TRANSCRIPT text T after its honesty gates + the streaming
+    // reconciliation. Everything the client should see has been emitted by then.
+    const content = await callClaudeTui(model, messages, conversationId, authInfo.keyName, res, streamCtx);
+    // Cache T — never the concatenated deltas (mirrors the buffered TUI path).
+    if (CACHE_TTL > 0 && authInfo.cacheHash) {
+      try { setCachedResponse(authInfo.cacheHash, model, content); } catch (e) { logEvent("error", "cache_write_failed", { error: e.message }); }
+    }
+    if (!res.writableEnded && !res.destroyed) {
+      sendSSE(res, {
+        id, object: "chat.completion.chunk", created, model,
+        choices: [{ index: 0, delta: {}, finish_reason: "stop" }],
+      }, hb);
+      res.write("data: [DONE]\n\n");
+      res.end();
+    }
+    try { recordUsage({ keyId: authInfo.keyId, keyName: authInfo.keyName, model, promptChars, responseChars: content.length, elapsedMs: Date.now() - t0, success: true }); } catch (e) { logEvent("error", "usage_record_failed", { error: e.message }); }
+  } catch (err) {
+    // Client walked away (queued OR mid-turn): nothing to write to, nothing to record —
+    // same quiet outcome as every other disconnect path (L1 / F2).
+    if (err instanceof RequestDisconnectedError) { try { res.end(); } catch {} return; }
+    try { recordUsage({ keyId: authInfo.keyId, keyName: authInfo.keyName, model, promptChars, responseChars: 0, elapsedMs: Date.now() - t0, success: false }); } catch (e) { logEvent("error", "usage_record_failed", { error: e.message }); }
+    console.error(`[proxy] error: ${err.message}`);
+    // Headers are already out (eager, above), so — exactly like the -p path — the failure is
+    // surfaced as an SSE error frame, NOT a success-looking finish_reason:"stop". This is what
+    // keeps a truncated turn, an auth banner, or a stream divergence from being served as an
+    // answer: the client sees an error, and nothing was cached.
+    if (!res.writableEnded && !res.destroyed) {
+      sendSSE(res, { error: { message: sanitizeError(err.message), type: "provider_error" } }, hb);
+      res.write("data: [DONE]\n\n");
+      res.end();
+    }
+  } finally {
+    hb.stop();
+    detach();
+  }
+}
+
 // ── SSE heartbeat (opt-in idle watchdog) ────────────────────────────────
 // Emits `: keepalive\n\n` SSE comment frames during silent windows on the
 // streaming response. Design: docs/superpowers/specs/2026-04-25-47-sse-heartbeat-design.md
@@ -2260,6 +2582,11 @@ async function handleChatCompletions(req, res) {
  }

  if (stream) {
+    if (TUI_MODE && TUI_STREAM) {
+      // TUI-mode REAL streaming (opt-in): emit delta.content chunks as claude renders them,
+      // via its MessageDisplay hook. The transcript remains authoritative (gates + cache).
+      return callClaudeTuiStreaming(model, messages, conversationId, res, { keyId: req._authKeyId, keyName: req._authKeyName, cacheHash: req._cacheHash });
+    }
    if (TUI_MODE) {
      // TUI-mode: no real token stream — buffer the full turn via callClaudeTui,
      // optionally write-back to cache, then replay as chunked SSE.
@@ -2558,9 +2885,17 @@ const server = createServer(async (req, res) => {
      // still appears with enabled:false (cheap, harmless) so the shape is stable.
      // entrypointMismatches/lastEntrypoint exist so an operator can poll /health to catch a
      // silent metered-pool drift (the audit's top risk after the 6/15 billing flip).
+      // `pool` is a NEW nested field inside the (already additive) tui block: null when the
+      // warm pool is off (the default), so the disabled shape is unchanged apart from one
+      // explicit null. Lets the operator confirm hit rate + standing process cost.
+      //
+      // streamEnabled + the stream* counters are likewise ADDITIVE (new fields only, same
+      // grandfathered B.2 rationale — ADR 0006). streamDivergences is the one an operator
+      // must watch: a non-zero value means a streamed turn was REFUSED because the deltas
+      // disagreed with the transcript, which is the streaming path's only correctness risk.
      tui: buildTuiHealthBlock(
-        { enabled: TUI_MODE, entrypointMode: TUI_ENTRYPOINT, maxConcurrent: TUI_MAX_CONCURRENT },
-        tuiStats, tuiSemaphore,
+        { enabled: TUI_MODE, entrypointMode: TUI_ENTRYPOINT, maxConcurrent: TUI_MAX_CONCURRENT, streamEnabled: TUI_MODE && TUI_STREAM },
+        tuiStats, tuiSemaphore, tuiPool,
      ),
    });
  }
@@ -2797,6 +3132,27 @@ function gracefulShutdown(signal) {
  if (tuiReapInterval) clearInterval(tuiReapInterval);
  closeDb();

+  // 2b. Drain the warm pane pool. A pooled `claude` is a child of the tmux SERVER, not of
+  // this node process, so it is NOT in activeProcesses and step 3 below cannot reach it —
+  // without this explicit drain every warm pane would outlive OCP as an orphan (and the
+  // pool's in-memory registry dies with the process, so nothing would remember it owned them).
+  //
+  // drain() kills the pane that is currently BOOTING too, and it does so SYNCHRONOUSLY. That
+  // is required, not incidental: step 4 below calls process.exit(0) in THIS SAME TICK whenever
+  // activeProcesses is empty — which on a TUI host it always is — so any cleanup a boot
+  // deferred to a .then()/.catch() would simply never run. (That was a real bug: the pool used
+  // to track in-flight boots as a count, could not name the booting session, and orphaned a
+  // live authenticated `claude` on every shutdown that landed mid-boot.)
+  //
+  // Orphans that survive anyway (SIGKILL, power loss) are still caught by the next instance's
+  // boot reap — this makes the graceful path clean, it is not the only safety net.
+  if (tuiPool) {
+    try {
+      const drained = tuiPool.drain();
+      if (drained) logEvent("info", "tui_pool_drained", { count: drained, trigger: "shutdown" });
+    } catch (e) { logEvent("error", "tui_pool_drain_failed", { error: e.message }); }
+  }
+
  // 3. Kill all active child processes
  for (const proc of activeProcesses) {
    try { proc.kill("SIGTERM"); } catch {}
@@ -2866,11 +3222,19 @@ server.listen(PORT, BIND_ADDRESS, () => {
      ? (TUI_HOME === process.env.HOME ? "env-token (real home — unset OCP_TUI_HOME for credential isolation)" : "env-token (credential-isolated home — no credentials.json)")
      : "credentials.json (no CLAUDE_CODE_OAUTH_TOKEN — see Troubleshooting #401)";
    console.log(`  TUI-mode: ON home=${TUI_HOME} cwd=${TUI_CWD} auth=${tuiAuth} wallclock=${TUI_WALLCLOCK_MS}ms maxConcurrent=${TUI_MAX_CONCURRENT}`);
+    console.log(TUI_POOL_SIZE > 0
+      ? `  TUI warm pool: ON size=${TUI_POOL_SIZE} — ${TUI_POOL_SIZE} idle \`claude\` process(es) held warm; first request per model is still a cold MISS`
+      : `  TUI warm pool: OFF (set OCP_TUI_POOL_SIZE=1..${POOL_MAX_SIZE} to pre-boot panes and cut ~3-4s per request)`);
    try {
      // F7 fix: scope to THIS instance's own port (see reapStaleTuiSessions). includeLegacy:
      // true ONLY here — the one-time boot reap is the designated point to claim orphaned
      // bare-prefix ("ocp-tui-<uuid8>") zombie sessions left by a PRE-fix process generation
      // of this same instance (no live post-fix instance ever creates that shape again).
+      // No `spare`: the warm pool is EMPTY at boot (there is no boot-time pre-warm — the pool
+      // learns its model from the first request), so this reap has no live pane to protect and
+      // it is exactly what SHOULD claim any ocp-tui-<port>-p* pool orphans left by a previous
+      // process generation of this instance (POOL/REAPER INVARIANT property 2). If a future
+      // change ever pre-warms at boot, this call MUST start passing tuiPool.liveNames().
      const n = reapStaleTuiSessions({ port: PORT, includeLegacy: true });
      if (n) logEvent("info", "tui_reaped_stale_sessions", { count: n });
    } catch {}
@@ -23,9 +23,30 @@ process.env.HOME = homedir(); // ensure consistent
 let passed = 0;
 let failed = 0;

+// Pending promises from tests declared `async` but registered through the SYNC `test()` helper.
+// 44 tests in this file are written that way. Before this, `test()` called fn(), got a promise back,
+// and immediately printed ✓ and incremented `passed` — WITHOUT AWAITING IT. So for every async test:
+//   - ✓ meant "did not throw synchronously", NOT "passed";
+//   - a failed assertion escaped as an unhandled rejection, which crashes the process (CI still goes
+//     red on the non-zero exit) but is NOT counted, so the summary could print "N passed, 0 failed"
+//     and be wrong.
+// The suite's own headline number was therefore not evidence for any async test — including the
+// regression guards in this PR. Collected here and awaited before the summary prints.
+const pendingAsync = [];
+
 function test(name, fn) {
  try {
-    fn();
+    const r = fn();
+    if (r && typeof r.then === "function") {
+      // Async body: settle it before counting. Do NOT print ✓ yet.
+      pendingAsync.push(
+        r.then(
+          () => { passed++; console.log(`  ✓ ${name}`); },
+          (e) => { failed++; console.log(`  ✗ ${name}: ${e.message}`); },
+        ),
+      );
+      return;
+    }
    passed++;
    console.log(`  ✓ ${name}`);
  } catch (e) {
@@ -1795,6 +1816,65 @@ test("buildTuiCmd shq-escapes a token containing shell metacharacters (no inject
  }
 });

+// OCP_TUI_EFFORT (TUI latency, docs/plans/2026-07-13-tui-latency): the pane's claude
+// must get an EXPLICIT --effort so its effort never depends on which HOME mode
+// resolveTuiHome() picked (real-home inherits the operator's settings.json effortLevel;
+// env-token scratch inherits claude's built-in default).
+test("buildTuiCmd passes --effort low by default (OCP_TUI_EFFORT unset)", () => {
+  const save = process.env.OCP_TUI_EFFORT;
+  try {
+    delete process.env.OCP_TUI_EFFORT;
+    const cmd = buildTuiCmd("/usr/bin/claude", "m", "sid-eff1", "/home/u", "cli");
+    assert.ok(cmd.includes("--effort low"), "default must pin --effort low");
+  } finally {
+    if (save === undefined) delete process.env.OCP_TUI_EFFORT;
+    else process.env.OCP_TUI_EFFORT = save;
+  }
+});
+
+test("buildTuiCmd honors an explicit OCP_TUI_EFFORT level (case/space-normalized)", () => {
+  const save = process.env.OCP_TUI_EFFORT;
+  try {
+    process.env.OCP_TUI_EFFORT = " XHigh ";
+    const cmd = buildTuiCmd("/usr/bin/claude", "m", "sid-eff2", "/home/u", "cli");
+    assert.ok(cmd.includes("--effort xhigh"), "explicit level must be passed, normalized");
+    assert.ok(!cmd.includes("--effort low"), "default must not also appear");
+  } finally {
+    if (save === undefined) delete process.env.OCP_TUI_EFFORT;
+    else process.env.OCP_TUI_EFFORT = save;
+  }
+});
+
+test("buildTuiCmd OCP_TUI_EFFORT=inherit omits --effort entirely (pre-flag argv)", () => {
+  const save = process.env.OCP_TUI_EFFORT;
+  try {
+    process.env.OCP_TUI_EFFORT = "inherit";
+    const cmd = buildTuiCmd("/usr/bin/claude", "m", "sid-eff3", "/home/u", "cli");
+    assert.ok(!/--effort/.test(cmd), "inherit must not add --effort");
+  } finally {
+    if (save === undefined) delete process.env.OCP_TUI_EFFORT;
+    else process.env.OCP_TUI_EFFORT = save;
+  }
+});
+
+test("buildTuiCmd falls back to --effort low on an invalid OCP_TUI_EFFORT (never reaches argv)", () => {
+  const save = process.env.OCP_TUI_EFFORT;
+  const savedErr = console.error;
+  try {
+    process.env.OCP_TUI_EFFORT = "ludicrous'; rm -rf /;'";
+    let warned = "";
+    console.error = (...a) => { warned = a.join(" "); };
+    const cmd = buildTuiCmd("/usr/bin/claude", "m", "sid-eff4", "/home/u", "cli");
+    assert.ok(cmd.includes("--effort low"), "invalid value must fall back to low");
+    assert.ok(!cmd.includes("ludicrous"), "invalid raw value must NOT reach the shell string");
+    assert.ok(/invalid OCP_TUI_EFFORT/.test(warned), "must log a warning");
+  } finally {
+    console.error = savedErr;
+    if (save === undefined) delete process.env.OCP_TUI_EFFORT;
+    else process.env.OCP_TUI_EFFORT = save;
+  }
+});
+
 test("buildTuiCmd OCP_TUI_FULL_TOOLS=1 grants -p-equivalent tool surface (single-user opt-in)", () => {
  const save = { ...process.env };
  const restore = () => {
@@ -1968,6 +2048,489 @@ test("reaper with includeLegacy=true still spares a sibling instance's port-scop
  assert.ok(!calls.includes("kill-server"), "sibling instance's live session still blocks kill-server");
 });

+// ── TUI warm pane pool (docs/plans/2026-07-13-tui-latency #3) ────────────
+import { TuiPanePool, resolvePoolSize, POOL_MAX_SIZE, POOL_MAX_AGE_MS } from "./lib/tui/pool.mjs";
+import { poolPaneName as poolName } from "./lib/tui/session.mjs";
+
+// A pool wired to fakes: no tmux, no claude. bootPane resolves on the microtask queue; use
+// `await settle()` after a refill() to let the SERIALIZED boot chain run to target.
+// `live` models the real tmux server: bootTuiPane creates the session SYNCHRONOUSLY and only
+// THEN waits (up to POOL_BOOT_MS) for the input bar, so the fake boot registers the session
+// immediately and only afterwards resolves. `opts.hold` keeps a boot in that mid-flight window
+// so tests can act on a pane that is live-but-not-yet-warm — the state that hid two bugs.
+function makeFakePool(opts = {}) {
+  const killed = [];
+  const booted = [];
+  const live = new Set();   // "tmux sessions" that currently exist
+  let seq = 0;
+  let clock = 1_000_000;
+  const healthy = new Set();
+  // FIFO gate queue — one entry per in-flight held boot. A single `release` slot would be
+  // OVERWRITTEN by a later boot, so releasing "the first boot" would silently release the
+  // second instead (and mask the stale-settle bug this harness exists to test).
+  const gates = [];
+  const pool = new TuiPanePool({
+    size: opts.size ?? 2,
+    maxAgeMs: opts.maxAgeMs ?? POOL_MAX_AGE_MS,
+    now: () => clock,
+    mintPane: () => {
+      const n = ++seq;
+      return { sessionId: `sid-${n}`, name: `ocp-tui-3456-p${String(n).padStart(8, "0")}` };
+    },
+    bootPane: async (model, { sessionId, name }) => {
+      live.add(name);                                     // session exists NOW
+      booted.push({ name, model });
+      if (opts.hold) {
+        await new Promise((r) => gates.push(r));          // ...stuck waiting for readiness
+      }
+      if (opts.bootThrows) { live.delete(name); throw new Error("boom"); }
+      // A pane whose session was killed while booting can never become ready — exactly what
+      // the real bootTuiPane does (it throws tui_pane_not_ready).
+      if (!live.has(name)) throw new Error("tui_pane_not_ready");
+      healthy.add(name);
+      return { name, sessionId, model, bootedAt: clock };
+    },
+    killPane: (name) => { killed.push(name); healthy.delete(name); live.delete(name); },
+    paneHealthy: (name) => healthy.has(name),
+  });
+  return {
+    pool, killed, booted, healthy, live,
+    releaseBoot: () => { const r = gates.shift(); if (r) r(); },  // release the OLDEST held boot
+    advance: (ms) => { clock += ms; }, at: () => clock,
+  };
+}
+const tick = () => new Promise((r) => setImmediate(r));
+// Refills are SERIALIZED (one boot at a time, re-kicked on success), so settling the pool
+// takes a chain of microtask turns, not one. 40 is far more than POOL_MAX_SIZE needs.
+const settle = async () => { for (let i = 0; i < 40; i++) await tick(); };
+
+console.log("\nTUI warm pane pool (acquire / miss / refill / TTL / reaper exemption):");
+
+test("resolvePoolSize: default/garbage/negative disable the pool; size is clamped to POOL_MAX_SIZE", () => {
+  assert.equal(resolvePoolSize(undefined), 0, "unset => off (byte-for-byte today's cold path)");
+  assert.equal(resolvePoolSize("0"), 0);
+  assert.equal(resolvePoolSize("-3"), 0);
+  assert.equal(resolvePoolSize("banana"), 0, "garbage disables rather than guessing a size");
+  assert.equal(resolvePoolSize("2"), 2);
+  assert.equal(resolvePoolSize("99"), POOL_MAX_SIZE, "clamped — never boot an unbounded number of idle claudes");
+});
+
+test("pool size 0 is inert: acquire always misses and refill never boots", async () => {
+  const { pool, booted } = makeFakePool({ size: 0 });
+  assert.equal(pool.enabled, false);
+  assert.equal(pool.acquire("m1"), null, "disabled pool always MISSES → caller cold-boots");
+  pool.refill();
+  await settle();
+  assert.equal(booted.length, 0, "a disabled pool must never spawn a process");
+});
+
+test("acquire MISSES on an empty pool, and the miss refills for the requested model", async () => {
+  const { pool, booted } = makeFakePool({ size: 2 });
+  assert.equal(pool.acquire("sonnet"), null, "first request is always a MISS (no boot-time pre-warm)");
+  assert.equal(pool.misses, 1);
+  pool.refill();
+  await settle();
+  assert.equal(pool.warm, 2, "refilled to target");
+  assert.deepEqual(booted.map((b) => b.model), ["sonnet", "sonnet"], "warmed for the model that missed");
+});
+
+test("acquire HITS a warm pane, hands it out ONCE, and never returns it (single-use)", async () => {
+  const { pool } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  assert.equal(pool.warm, 2);
+
+  const a = pool.acquire("sonnet");
+  assert.ok(a && a.name && a.sessionId, "warm pane handed out");
+  assert.equal(pool.hits, 1);
+  assert.equal(pool.warm, 1, "the pane LEAVES the registry when acquired");
+
+  const b = pool.acquire("sonnet");
+  assert.notEqual(b.name, a.name, "a pane is NEVER handed out twice — single-use");
+  assert.notEqual(b.sessionId, a.sessionId, "each pane carries its OWN fresh session-id (transcript.mjs scoping)");
+  assert.equal(pool.warm, 0);
+  assert.equal(pool.acquire("sonnet"), null, "exhausted pool MISSES rather than reusing a pane");
+});
+
+test("refill is bounded: never more than `size` panes, and concurrent refills do not overshoot", async () => {
+  const { pool, booted } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet");
+  pool.refill(); pool.refill(); pool.refill(); // hammer it
+  await settle();
+  assert.equal(pool.warm, 2, "still exactly `size` warm panes");
+  assert.equal(booted.length, 2, "the _booting guard prevented duplicate boots");
+});
+
+// Live finding at size=2: two cold `claude` boots racing an in-flight turn made a refill
+// overrun even the generous pool readiness cap. Boots are therefore SERIALIZED.
+test("refill boots panes ONE AT A TIME (never two claude cold-boots racing each other)", async () => {
+  let concurrent = 0, peak = 0;
+  let seq = 0;
+  const pool = new TuiPanePool({
+    size: 3,
+    mintPane: () => { const n = ++seq; return { sessionId: `s${n}`, name: `p${n}` }; },
+    bootPane: async (model, { sessionId, name }) => {
+      concurrent++; peak = Math.max(peak, concurrent);
+      await new Promise((r) => setImmediate(r)); // simulate boot latency
+      concurrent--;
+      return { name, sessionId, model, bootedAt: Date.now() };
+    },
+    killPane: () => {},
+    paneHealthy: () => true,
+  });
+  pool.acquire("sonnet");
+  pool.refill();
+  await settle();
+  assert.equal(pool.warm, 3, "chain still reaches the target size");
+  assert.equal(peak, 1, "at most ONE boot in flight at any moment");
+});
+
+test("a FAILED boot does not re-kick the chain (backoff — a broken claude must not spin)", async () => {
+  const { pool, booted } = makeFakePool({ size: 3, bootThrows: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await settle();
+  assert.equal(pool.bootFailures, 1, "counted as a genuine failure (nobody cancelled it)");
+  assert.equal(booted.length, 1, "exactly ONE attempt — a failure stops the chain, it does not respawn forever");
+  assert.equal(pool.warm, 0);
+  assert.equal(pool.booting, 0, "and the booting slot is released, so the next trigger can retry");
+});
+
+test("acquire drops an UNHEALTHY warm pane (kills it) and falls through to a MISS", async () => {
+  const { pool, killed, healthy, booted } = makeFakePool({ size: 1 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  const dead = booted[0].name;
+  healthy.delete(dead); // pane died / stopped being input-ready while idle
+
+  assert.equal(pool.acquire("sonnet"), null, "a dead pane must MISS, never hang a turn");
+  assert.ok(killed.includes(dead), "the dead pane is killed, not leaked");
+  assert.equal(pool.misses, 2);
+});
+
+test("acquire drops an EXPIRED warm pane (older than maxAgeMs)", async () => {
+  const { pool, killed, booted, advance } = makeFakePool({ size: 1, maxAgeMs: 60_000 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  advance(60_001);
+  assert.equal(pool.acquire("sonnet"), null, "a pane past its TTL is not handed out");
+  assert.ok(killed.includes(booted[0].name), "expired pane is killed");
+});
+
+test("a model switch drops the wrong-model panes and retargets the pool (--model is fixed at spawn)", async () => {
+  const { pool, killed, booted } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  const sonnetPanes = booted.map((b) => b.name);
+
+  assert.equal(pool.acquire("opus"), null, "different model => MISS (a sonnet pane cannot serve opus)");
+  assert.equal(pool.warm, 0, "sonnet panes dropped");
+  for (const p of sonnetPanes) assert.ok(killed.includes(p), "wrong-model pane killed, not leaked");
+  assert.equal(pool.warmModel, "opus", "pool retargeted to the model actually being asked for");
+
+  pool.refill(); await settle();
+  assert.deepEqual(booted.slice(2).map((b) => b.model), ["opus", "opus"], "refilled for the NEW model");
+});
+
+test("a boot that resolves AFTER a drain kills its own pane instead of enlisting it", async () => {
+  const { pool, live } = makeFakePool({ size: 1 });
+  pool.acquire("sonnet");
+  pool.refill();          // boot is in flight...
+  pool.drain();           // ...pool drained before it resolves (shutdown / reap sweep)
+  await settle();
+  assert.equal(pool.warm, 0, "the late pane must NOT be enlisted into a drained pool");
+  // Assert LIVENESS, not the kill-call COUNT. This assertion used to read
+  // `assert.equal(killed.length, 1)` — and it PASSED while the orphan it is named after was
+  // actually present: _cancelBooting kills BY NAME, and at drain time the tmux session does not
+  // exist yet (bootPane runs on a microtask), so that kill is a NO-OP which still increments the
+  // counter. "kill was called once" and "a live session is orphaned" were both true at the same
+  // time. The only honest question is whether the session is dead.
+  assert.equal(live.size, 0, "it kills itself — no orphan process left behind");
+});
+
+test("bootPane failure is counted, never thrown into the request path, and does not wedge refill", async () => {
+  const { pool } = makeFakePool({ size: 1, bootThrows: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await settle();
+  assert.equal(pool.warm, 0);
+  assert.equal(pool.bootFailures, 1);
+  assert.equal(pool.booting, 0, "the _booting counter is released on failure (else refill wedges forever)");
+  assert.equal(pool.acquire("sonnet"), null, "and the caller just MISSES → cold path");
+});
+
+test("drain kills every warm pane and pauses refills; resume restarts them", async () => {
+  const { pool, killed } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  assert.equal(pool.warm, 2);
+
+  assert.equal(pool.drain(), 2, "drain reports how many it killed");
+  assert.equal(pool.warm, 0);
+  assert.equal(killed.length, 2, "both panes killed — none outlive the drain");
+
+  pool.refill(); await settle();
+  assert.equal(pool.warm, 0, "refill is a NO-OP while drained (paused)");
+
+  pool.resume(); await settle();
+  assert.equal(pool.warm, 2, "resume refills");
+});
+
+// ── The crux: pool ↔ reaper coexistence (POOL/REAPER INVARIANT, lib/tui/session.mjs) ──
+console.log("\nTUI warm pool ↔ session reaper coexistence:");
+
+test("INVARIANT 1: a LIVE pooled pane is NEVER reaped (it is in the spare set)", () => {
+  const killed = [];
+  const live = "ocp-tui-3456-pdeadbeef";
+  const fakeTmux = (args) => {
+    if (args[0] === "list-sessions") return { status: 0, stdout: `${live}\nocp-tui-3456-aaaa\n` };
+    if (args[0] === "kill-session") { killed.push(args[args.indexOf("-t") + 1]); return { status: 0 }; }
+    return { status: 0, stdout: "" };
+  };
+  const n = reapStaleTuiSessions({ tmux: fakeTmux, port: 3456, spare: new Set([live]) });
+  assert.equal(n, 1, "only the stale turn session was reaped");
+  assert.ok(!killed.includes(live), "the live warm pane must survive the sweep");
+  assert.ok(killed.includes("ocp-tui-3456-aaaa"), "a genuinely stale own session is still reaped");
+});
+
+test("INVARIANT 2: an ORPHANED pooled pane (pool-shaped but NOT in the spare set) IS reaped", () => {
+  const killed = [];
+  // ocp-tui-3456-porphan01 LOOKS pooled but the live registry does not claim it — e.g. left
+  // behind by a previous process generation, whose in-memory registry died with it.
+  const fakeTmux = (args) => {
+    if (args[0] === "list-sessions") return { status: 0, stdout: "ocp-tui-3456-porphan01\nocp-tui-3456-plive0001\n" };
+    if (args[0] === "kill-session") { killed.push(args[args.indexOf("-t") + 1]); return { status: 0 }; }
+    return { status: 0, stdout: "" };
+  };
+  const n = reapStaleTuiSessions({ tmux: fakeTmux, port: 3456, spare: new Set(["ocp-tui-3456-plive0001"]) });
+  assert.equal(n, 1);
+  assert.deepEqual(killed, ["ocp-tui-3456-porphan01"], "exemption is by EXACT NAME, never by name shape");
+});
+
+test("INVARIANT 2b: with NO spare set (the pre-pool call shape) pool-shaped panes are reaped — fail-safe", () => {
+  const killed = [];
+  const fakeTmux = (args) => {
+    if (args[0] === "list-sessions") return { status: 0, stdout: "ocp-tui-3456-pdeadbeef\n" };
+    if (args[0] === "kill-session") { killed.push(args[args.indexOf("-t") + 1]); return { status: 0 }; }
+    return { status: 0, stdout: "" };
+  };
+  const n = reapStaleTuiSessions({ tmux: fakeTmux, port: 3456 });
+  assert.equal(n, 1, "omitting `spare` reaps MORE, never less — forgetting it can't leak panes");
+  assert.deepEqual(killed, ["ocp-tui-3456-pdeadbeef"]);
+});
+
+test("INVARIANT 3: kill-server is SUPPRESSED while a live pooled pane is spared", () => {
+  const calls = [];
+  const live = "ocp-tui-3456-plive0001";
+  const fakeTmux = (args) => {
+    calls.push(args.join(" "));
+    if (args[0] === "list-sessions") return { status: 0, stdout: `${live}\nocp-tui-3456-aaaa\n` };
+    return { status: 0, stdout: "" };
+  };
+  reapStaleTuiSessions({ tmux: fakeTmux, port: 3456, spare: new Set([live]) });
+  assert.ok(!calls.includes("kill-server"), "kill-server would kill the live pane (a child of the tmux server)");
+});
+
+test("INVARIANT 3b: after a DRAIN the spare set is empty, so kill-server fires again (zombie reaping preserved)", async () => {
+  // This is the whole reason server.mjs drains BEFORE the periodic sweep: a permanently-full
+  // pool would otherwise permanently suppress the only mechanism that reaps defunct claudes.
+  const { pool } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  assert.equal(pool.liveNames().size, 2, "pool is full → the sweep would be suppressed");
+
+  pool.drain();
+  assert.equal(pool.liveNames().size, 0, "drain empties the live registry");
+
+  const calls = [];
+  const fakeTmux = (args) => {
+    calls.push(args.join(" "));
+    if (args[0] === "list-sessions") return { status: 0, stdout: "ocp-tui-3456-aaaa\n" };
+    return { status: 0, stdout: "" };
+  };
+  reapStaleTuiSessions({ tmux: fakeTmux, port: 3456, spare: pool.liveNames() });
+  assert.ok(calls.includes("kill-server"), "kill-server fires post-drain — defunct zombies still get reaped");
+});
+
+// ── MID-BOOT: the state that hid M1a + M1b ────────────────────────────────────────────
+// bootTuiPane creates the tmux session SYNCHRONOUSLY, then waits up to POOL_BOOT_MS (20s)
+// for the input bar. So a pooled session can be LIVE for ~20s before its boot resolves.
+// Every reaper test above uses a pool that is either full or drained — never mid-boot.
+// That gap is exactly why both bugs shipped past the first round of tests.
+
+test("M1a: a reap tick during an IN-FLIGHT BOOT must not orphan-kill the booting pane", async () => {
+  const { pool, live } = makeFakePool({ size: 1, hold: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await tick();                                    // boot started; session live; NOT yet warm
+  assert.equal(pool.warm, 0, "not warm yet");
+  assert.equal(pool.booting, 1, "a boot is in flight");
+  assert.equal(live.size, 1, "...and its tmux session ALREADY EXISTS");
+
+  const bootingName = [...live][0];
+  assert.ok(pool.liveNames().has(bootingName),
+    "REGRESSION GUARD: the booting pane MUST be nameable, or the sweep cannot spare it " +
+    "(the pool used to track in-flight boots as a COUNT and this was empty)");
+});
+
+test("M1a: the reap tick's drain kills the booting pane, and resume() starts a FRESH boot", async () => {
+  const warns = [];
+  const { pool, live, releaseBoot } = makeFakePool({ size: 1, hold: true });
+  pool._log = (lvl, ev) => { if (lvl === "warn") warns.push(ev); };
+  pool.acquire("sonnet");
+  pool.refill();
+  await tick();
+  const first = [...live][0];
+
+  // The reap tick, as server.mjs runs it: drain -> reap -> resume.
+  const drained = pool.drain();
+  assert.equal(drained, 1, "drain accounts for the booting pane");
+  assert.equal(live.size, 0, "its tmux session is killed — kill-server can now flush zombies");
+  assert.equal(pool.liveNames().size, 0, "nothing left to spare, so kill-server is not suppressed");
+
+  pool.resume();
+  await tick();
+  assert.equal(pool.booting, 1, "resume() started a FRESH boot — the pool is not left empty with nothing scheduled");
+  assert.notEqual([...live][0], first, "and it is a NEW pane, not the killed one");
+
+  // Now let the ORIGINAL (cancelled) boot settle. It rejects with tui_pane_not_ready because
+  // we killed its session — but that is OUR doing, not a fault.
+  releaseBoot();
+  await settle();
+  assert.equal(pool.bootFailures, 0,
+    "a cancelled boot must NOT be counted as a bootFailure — that is the WARN operators alert on");
+  assert.deepEqual(warns, [], "and it must not log tui_pool_boot_failed for a healthy drain");
+  assert.equal(pool.cancelled, 1, "it is counted as a cancellation instead (counted exactly once)");
+});
+
+test("M1b: shutdown drain kills the booting pane SYNCHRONOUSLY — no orphaned claude", async () => {
+  const { pool, live } = makeFakePool({ size: 1, hold: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await tick();
+  assert.equal(live.size, 1, "a live pooled session exists");
+
+  // gracefulShutdown: drain() then process.exit(0) IN THE SAME TICK (TUI panes are tmux
+  // children, not node children, so activeProcesses is empty and the exit is immediate).
+  // Nothing scheduled on the microtask queue can run. So we assert WITHOUT awaiting.
+  pool.drain();
+  assert.equal(live.size, 0,
+    "REGRESSION GUARD: the pane must be dead BEFORE any await. A .then()-based cleanup would " +
+    "never run before process.exit and would orphan a live authenticated `claude`.");
+});
+
+// M1b, second costume: drain() in the SAME synchronous block as refill(). The tmux session does
+// not exist yet at drain time (bootPane runs on a microtask), so _cancelBooting's kill-by-name is
+// a no-op — and a `.then` that merely `return`s on a stale generation would then let the boot
+// CREATE the session and walk away from it. Not reachable from any current call site, but ADR 0008
+// and the reap-tick comment both contemplate a boot-time pre-warm, which is exactly this shape.
+// NOTE: deliberately NOT `hold: true`. A held boot never settles, so its `.then` never runs and
+// the guard would vacuously pass — the test must let the boot actually SUCCEED, because the bug is
+// precisely that a SUCCESSFUL boot on a cancelled generation walks away from its live session.
+test("M1b': a boot cancelled BEFORE its session existed is still killed when it settles", async () => {
+  const { pool, live } = makeFakePool({ size: 1 });
+  pool.acquire("sonnet");                 // miss → learns the model
+
+  pool.refill();   // mints the identity; bootPane is queued on a microtask — no session YET
+  pool.drain();    // SAME sync block: kill-by-name finds nothing to kill (no-op), bumps the gen
+  assert.equal(live.size, 0, "precondition: the session genuinely did not exist at cancel time");
+
+  await tick();    // NOW the boot microtask runs, CREATES the session, and settles on a stale gen
+  await tick();
+
+  assert.equal(live.size, 0,
+    "REGRESSION GUARD: a stale-generation boot must KILL its pane, not assume _cancelBooting " +
+    "already did. _cancelBooting kills BY NAME, and the tmux session does not exist until the " +
+    "boot microtask runs — so a cancellation landing first is a no-op, and a bare `return` here " +
+    "orphans a live authenticated `claude` that nothing owns.");
+});
+
+test("a stale boot settling after drain+resume must not clear the NEW boot's slot", async () => {
+  const { pool, releaseBoot } = makeFakePool({ size: 1, hold: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await tick();
+  pool.drain();                 // cancels boot #1 (generation bumped)
+  pool.resume();
+  await tick();
+  assert.equal(pool.booting, 1, "boot #2 owns the slot");
+  releaseBoot();                // boot #1 finally settles (rejects)
+  await settle();
+  assert.equal(pool.booting, 1, "boot #2 STILL owns the slot — a stale settle must not free it");
+});
+
+test("a model switch cancels an in-flight boot for the OLD model (kills it, frees the slot)", async () => {
+  const { pool, live, killed } = makeFakePool({ size: 1, hold: true });
+  pool.acquire("sonnet");
+  pool.refill();
+  await tick();
+  const sonnetPane = [...live][0];
+
+  pool.acquire("opus");         // retarget mid-boot
+  assert.ok(killed.includes(sonnetPane), "the old model's booting pane is killed, not left to linger");
+  assert.equal(pool.booting, 0, "and its slot is freed immediately, so the new model can boot now");
+  assert.equal(pool.warmModel, "opus");
+
+  pool.refill();
+  await tick();
+  assert.equal(pool.booting, 1, "a boot for the NEW model starts without waiting out the old one");
+});
+
+test("a pane handed out for a turn leaves the spare set immediately (so its teardown is authoritative)", async () => {
+  const { pool } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  const taken = pool.acquire("sonnet");
+  assert.ok(!pool.liveNames().has(taken.name),
+    "an acquired pane is the CALLER's — the pool must not also claim it live, or a crashed turn's pane would be spared forever");
+});
+
+test("N1: the pool mints ONE identity — the tmux name's hex is the transcript session-id's hex", async () => {
+  // Without this, `tmux ls` shows a pane whose name has no relation to any transcript file,
+  // so a live pane cannot be correlated to <HOME>/.claude/projects/*/<sessionId>.jsonl.
+  const seen = [];
+  const pool = new TuiPanePool({
+    size: 1,
+    mintPane: () => {
+      const sessionId = "deadbeef-1111-2222-3333-444444444444";
+      return { sessionId, name: poolName(3456, sessionId) };
+    },
+    bootPane: async (model, ident) => {
+      seen.push(ident);
+      return { ...ident, model, bootedAt: Date.now() };
+    },
+    killPane: () => {},
+    paneHealthy: () => true,
+  });
+  pool.acquire("sonnet");
+  pool.refill();
+  await settle();
+  assert.equal(seen.length, 1, "bootPane received the pool-minted identity");
+  assert.equal(seen[0].name, "ocp-tui-3456-pdeadbeef");
+  assert.ok(seen[0].name.endsWith(seen[0].sessionId.slice(0, 8)),
+    "the tmux session name carries the session-id's own hex — `tmux ls` correlates to the transcript");
+  const pane = pool.acquire("sonnet");
+  assert.equal(pane.sessionId, seen[0].sessionId,
+    "and the turn reads the transcript under THAT session-id — one identity end to end");
+});
+
+test("pool pane names are port-scoped (reapable as ours) and never match the legacy shape", () => {
+  const name = poolName(3456, "deadbeef-1111-2222-3333-444444444444");
+  assert.ok(name.startsWith(sessionPrefixForPort(3456)), "pool panes are OURS → reapable when orphaned");
+  assert.equal(name, "ocp-tui-3456-pdeadbeef");
+  assert.ok(!LEGACY_SESSION_NAME_RE.test(name), "must never be mistaken for a legacy bare-prefix session");
+  assert.ok(!poolName(9999, "aaaaaaaa-0000-0000-0000-000000000000").startsWith(sessionPrefixForPort(3456)),
+    "a sibling instance's pool pane is foreign to us");
+});
+
+test("buildTuiHealthBlock reports pool:null when off, and the pool's stats when on", async () => {
+  const st = { lastEntrypoint: "cli", entrypointMismatches: 0 };
+  const sem = { inflight: 0, queued: 0 };
+  const off = buildTuiHealthBlock({ enabled: true, entrypointMode: "cli", maxConcurrent: 2 }, st, sem, null);
+  assert.equal(off.pool, null, "pool disabled → explicit null (stable /health shape)");
+
+  const { pool } = makeFakePool({ size: 2 });
+  pool.acquire("sonnet"); pool.refill(); await settle();
+  const on = buildTuiHealthBlock({ enabled: true, entrypointMode: "cli", maxConcurrent: 2 }, st, sem, pool);
+  assert.equal(on.pool.size, 2);
+  assert.equal(on.pool.warm, 2);
+  assert.equal(on.pool.misses, 1);
+  assert.equal(on.pool.model, "sonnet");
+});
+
 // ── TUI home preparation (scratch vs real) ───────────────────────────────
 import { prepareTuiHome, ensureTuiCwdTrusted } from "./lib/tui/session.mjs";
 import { mkdtempSync as hMkdtemp, mkdirSync as hMkdir, writeFileSync as hWrite, readFileSync as hRead, existsSync as hExists, readlinkSync as hReadlink } from "node:fs";
@@ -2415,8 +2978,19 @@ test("buildTuiHealthBlock: shape + live counters (the additive /health tui block
  const ts = { lastEntrypoint: "cli", entrypointMismatches: 3 };
  const block = buildTuiHealthBlock(
    { enabled: true, entrypointMode: "cli", maxConcurrent: 2 }, ts, sem);
-  assert.deepEqual(Object.keys(block).sort(),
-    ["enabled", "entrypointMismatches", "entrypointMode", "inflight", "lastEntrypoint", "maxConcurrent", "queued"]);
+  // Shape is ADDITIVE-only: the seven original keys must all still be present (existing
+  // /health consumers are grandfathered, ADR 0006), plus `pool` (warm pane pool) and the
+  // stream* fields (backlog #2). Asserting CONTAINMENT plus an exact added-set — rather than
+  // one flat deepEqual — is what makes "additive" itself the thing under test: a future field
+  // that silently REPLACED an original key would pass a flat equality check that was updated
+  // alongside it, but cannot pass this one.
+  const ORIGINAL_KEYS = ["enabled", "entrypointMismatches", "entrypointMode", "inflight", "lastEntrypoint", "maxConcurrent", "queued"];
+  const keys = Object.keys(block);
+  for (const k of ORIGINAL_KEYS) assert.ok(keys.includes(k), `original /health key must survive: ${k}`);
+  assert.deepEqual(keys.filter((k) => !ORIGINAL_KEYS.includes(k)).sort(),
+    ["pool", "streamDeltas", "streamDivergences", "streamEnabled", "streamTopUps", "streamTurns", "streamZeroDeltaTurns"],
+    "only the documented pool + streaming fields may be added");
+  assert.equal(block.pool, null, "no pool passed → null (the default, pool disabled)");
  assert.equal(block.enabled, true);
  assert.equal(block.entrypointMode, "cli");
  assert.equal(block.lastEntrypoint, "cli");
@@ -2874,8 +3448,370 @@ async function runAsyncTests() {
  });
 }

+// ── TUI real streaming: MessageDisplay hook sink (backlog #2) ───────────────
+// Pure-logic coverage for lib/tui/stream.mjs: sink parsing, the concat===T assertion,
+// prefix-stability, the auth-banner holdback, message scoping, and the error paths.
+import { TuiDeltaAssembler, parseDeltaChunk, buildStreamSettings, streamFilePath, HOOK_SCRIPT, prepareStreamHook } from "./lib/tui/stream.mjs";
+
+test("stream: parseDeltaChunk consumes only COMPLETE lines (a torn write stays unread)", () => {
+  const p = (i, d, final = false) => JSON.stringify({ hook_event_name: "MessageDisplay", session_id: "s", message_id: "m", index: i, final, delta: d });
+  // second payload is mid-write — no trailing newline yet
+  const partial = `${p(0, "## A\n\n")}\n${p(1, "body").slice(0, 20)}`;
+  const r1 = parseDeltaChunk(partial, 0);
+  assert.equal(r1.deltas.length, 1, "only the terminated line is consumed");
+  assert.equal(r1.consumed, 1);
+  // now it lands complete
+  const whole = `${p(0, "## A\n\n")}\n${p(1, "body")}\n`;
+  const r2 = parseDeltaChunk(whole, r1.consumed);
+  assert.equal(r2.deltas.length, 1, "the once-partial line is picked up exactly once");
+  assert.equal(r2.deltas[0].delta, "body");
+  assert.equal(r2.consumed, 2);
+  // idempotent: nothing new
+  assert.equal(parseDeltaChunk(whole, r2.consumed).deltas.length, 0);
+});
+
+test("stream: parseDeltaChunk skips blank/garbage lines and foreign hook events", () => {
+  const md = JSON.stringify({ hook_event_name: "MessageDisplay", message_id: "m", index: 0, final: true, delta: "ok" });
+  const other = JSON.stringify({ hook_event_name: "Stop", message_id: "m", delta: "nope" });
+  const text = `\n{not json\n${other}\n${md}\n`;
+  const { deltas } = parseDeltaChunk(text, 0);
+  assert.equal(deltas.length, 1);
+  assert.equal(deltas[0].delta, "ok");
+});
+
+// The live-verified contract (claude 2.1.207): deltas are the raw markdown source and
+// concat(deltas) === extractLatestAssistantText(transcript), byte-exactly.
+const mdFire = (i, delta, { final = false, mid = "m1" } = {}) =>
+  ({ hook_event_name: "MessageDisplay", session_id: "s1", message_id: mid, index: i, final, delta });
+
+test("stream: concat(deltas) === T → exact, no top-up, prefix-stable at every n", () => {
+  const chunks = ["## Mutex\n\n", "A **mutual exclusion lock** prevents concurrent access.\n\n", "```javascript\nconst m = new Mutex();\n```"];
+  const T = chunks.join("");
+  const a = new TuiDeltaAssembler({ holdbackChars: 10 });
+  let acc = "";
+  chunks.forEach((c, i) => {
+    const out = a.push(mdFire(i, c, { final: i === chunks.length - 1 }));
+    if (out) acc += out;
+    assert.ok(T.startsWith(a.full), `prefix-stable at n=${i}`);
+  });
+  const rec = a.finalize(T);
+  assert.equal(rec.ok, true);
+  assert.equal(rec.exact, true, "concat(deltas) === T");
+  assert.equal(acc + rec.tail, T, "client's assembled stream === T");
+  assert.equal(a.deltas, 3);
+});
+
+test("stream: holdback withholds the first chars so the auth-banner gate can still fire", () => {
+  const banner = "Please run /login · API Error: 401 Invalid authentication credentials"; // 69 chars, a real one
+  const a = new TuiDeltaAssembler(); // default holdback 100
+  const out = a.push(mdFire(0, banner, { final: true }));
+  assert.equal(out, null, "a banner-length message must NEVER reach the client");
+  assert.equal(a.emitted, "", "nothing emitted");
+  // and the whole-message detector still classifies it — the gate runs on T, before any flush
+  assert.ok(detectTuiUpstreamError(a.full) !== null, "banner still detected at terminal");
+});
+
+test("stream: holdback releases once past the detector's reach, and only then", () => {
+  const a = new TuiDeltaAssembler({ holdbackChars: 100 });
+  assert.equal(a.push(mdFire(0, "x".repeat(80))), null, "80 chars: still held");
+  const out = a.push(mdFire(1, "y".repeat(40)));
+  assert.equal(out, "x".repeat(80) + "y".repeat(40), "released as one chunk once >100");
+  assert.equal(a.push(mdFire(2, "tail")), "tail", "subsequent deltas stream straight through");
+});
+
+test("stream: a short answer never passes the holdback and is delivered whole at terminal", () => {
+  const T = "The capital of France is Paris.";
+  const a = new TuiDeltaAssembler();
+  assert.equal(a.push(mdFire(0, T, { final: true })), null);
+  const rec = a.finalize(T);
+  assert.equal(rec.ok, true);
+  assert.equal(rec.exact, true);
+  assert.equal(rec.tail, T, "the whole short answer is flushed at terminal (buffered semantics)");
+});
+
+test("stream: a DROPPED delta is a safe prefix → top-up from the transcript, exact=false", () => {
+  const a = new TuiDeltaAssembler({ holdbackChars: 5 });
+  a.push(mdFire(0, "Hello world, this is the first block. "));
+  const T = "Hello world, this is the first block. And the tail the hook never delivered.";
+  const rec = a.finalize(T);
+  assert.equal(rec.ok, true, "prefix → recoverable");
+  assert.equal(rec.exact, false, "flagged: concat(deltas) !== T");
+  assert.equal(rec.tail, "And the tail the hook never delivered.");
+  assert.equal(a.emitted + rec.tail, T, "client still receives exactly T");
+});
+
+test("stream: emitted bytes NOT a prefix of T → divergence, refuse the turn", () => {
+  const a = new TuiDeltaAssembler({ holdbackChars: 5 });
+  a.push(mdFire(0, "Let me go and read that file for you first."));
+  const rec = a.finalize("A completely different final answer.");
+  assert.equal(rec.ok, false, "must NOT serve text the transcript disagrees with");
+  assert.equal(rec.tail, null);
+});
+
+// Message scoping: the transcript keeps only the LAST assistant message, so the assembler
+// must too. Discarding is safe while nothing has been emitted; after that it is a divergence.
+test("stream: new message_id BEFORE any emit → held text discarded, stays exact vs T", () => {
+  const a = new TuiDeltaAssembler({ holdbackChars: 100 });
+  a.push(mdFire(0, "I'll check the file.", { mid: "m1" })); // short pre-tool prose, held back
+  assert.equal(a.emitted, "");
+  const answer = "The file defines a Mutex class with acquire and release, and " + "z".repeat(90);
+  const out = a.push(mdFire(0, answer, { mid: "m2", final: true }));
+  assert.equal(out, answer, "only the FINAL message's text is emitted");
+  const rec = a.finalize(answer); // T = extractLatestAssistantText = the last message only
+  assert.equal(rec.ok, true);
+  assert.equal(rec.exact, true, "scoping to the last message_id keeps concat === T true");
+  assert.equal(a.messages, 2);
+});
+
+test("stream: new message_id AFTER an emit → unretractable, flagged and refused", () => {
+  const a = new TuiDeltaAssembler({ holdbackChars: 10 });
+  a.push(mdFire(0, "Long pre-tool prose that already went out to the client.", { mid: "m1" }));
+  assert.notEqual(a.emitted, "");
+  // Assert what push() RETURNS, not merely that finalize() refuses. This test used to check
+  // only restartedAfterEmit + finalize().ok, which left it passing while F1 was live: the
+  // second message's bytes were still being handed to the client. "The turn is refused" and
+  // "the client got the bytes anyway" were both true at once.
+  const out = a.push(mdFire(0, "The real answer.", { mid: "m2", final: true }));
+  assert.equal(out, null, "after a message boundary follows an emit, NOTHING more may be emitted");
+  assert.equal(a.restartedAfterEmit, true);
+  assert.equal(a.finalize("The real answer.").ok, false, "must refuse: prose already emitted is not in T");
+});
+
+test("F1: an auth banner rendered as a LATER message is never forwarded to the client", () => {
+  // The leak this class exists to prevent, in the shape production actually runs
+  // (OCP_TUI_FULL_TOOLS=1 → multi-message tool-using turns are the norm):
+  //   1. the model narrates past the holdback before a tool call  → released, emitted != ""
+  //   2. credentials expire mid-turn → claude renders the 401 as ordinary assistant TEXT,
+  //      as a NEW message
+  //   3. pre-fix: push() took the `if (this.released)` branch — `released` was never reset at
+  //      a message boundary — and returned the BANNER verbatim, straight to the client.
+  // The holdback protected only the FIRST message of a turn. This asserts it protects the rest.
+  const a = new TuiDeltaAssembler({ holdbackChars: 100 });
+  const narration = "I'll check that file for you and then report back with what I find inside it.";
+  a.push(mdFire(0, narration + narration, { mid: "m1" }));   // > holdback → released
+  assert.notEqual(a.emitted, "", "precondition: the narration really did reach the client");
+
+  const BANNER = "Please run /login · API Error: 401 Invalid authentication credentials";
+  const out = a.push(mdFire(1, BANNER, { mid: "m2", final: true }));
+  assert.equal(out, null, "the auth banner must NOT be forwarded once a later message begins");
+  assert.ok(!a.emitted.includes("401"), "no byte of the banner may have reached the client");
+  assert.equal(a.finalize(BANNER).ok, false, "and the turn is refused, not served");
+});
+
+test("F1: a first payload with message_id:null cannot disarm the guard", () => {
+  // The residual bypass the reviewer found by probing. `this.messageId` used to be initialized
+  // to null, so a first payload carrying message_id:null compared EQUAL to it → no boundary
+  // registered → `messages` stayed 0 → when the REAL boundary arrived, `messages > 1` evaluated
+  // 1 > 1 === false → restartedAfterEmit never armed → the released branch forwarded the banner.
+  // The whole F1 guard was disarmed by a single null field. parseDeltaChunk does not validate
+  // message_id, so such a payload does reach push().
+  const a = new TuiDeltaAssembler({ holdbackChars: 100 });
+  const narration = "I'll check that file for you and then report back with what I find inside it.";
+  a.push({ hook_event_name: "MessageDisplay", message_id: null, delta: narration + narration });
+  assert.notEqual(a.emitted, "", "precondition: the narration released to the client");
+  assert.equal(a.messages, 1, "a null message_id is still a MESSAGE — it must register as one");
+
+  const BANNER = "Please run /login · API Error: 401 Invalid authentication credentials";
+  const out = a.push({ hook_event_name: "MessageDisplay", message_id: "m2", delta: BANNER });
+  assert.equal(out, null, "the banner must not be forwarded — the guard must arm regardless");
+  assert.equal(a.restartedAfterEmit, true);
+  assert.ok(!a.emitted.includes("401"));
+});
+
+test("F1: whitespace cannot buy a release — the holdback screens TRIMMED length", () => {
+  // detectTuiUpstreamError() TRIMS before applying its <=100-char rule, so gating release on
+  // the UNTRIMMED pending.length let 101 spaces trim to "" → the detector has nothing to
+  // classify → returns null → release fires having screened nothing, and every subsequent
+  // delta of that message (a banner included) streams unfiltered.
+  const a = new TuiDeltaAssembler({ holdbackChars: 100 });
+  assert.equal(a.push(mdFire(0, " ".repeat(101), { mid: "m1" })), null,
+    "101 chars of whitespace must not clear a 100-char holdback");
+  assert.equal(a.released, false, "…and must not flip the assembler into released state");
+  const BANNER = "Please run /login · API Error: 401 Invalid authentication credentials";
+  assert.equal(a.push(mdFire(1, BANNER, { mid: "m1" })), null, "so the banner stays held back");
+  assert.ok(!a.emitted.includes("401"));
+});
+
+test("F3: a STALE or truncated hook script is overwritten, not trusted because it exists", () => {
+  // ~/.ocp-tui/stream/{md-hook.sh,settings.json} persist across OCP restarts. The old
+  // write-if-missing guard meant a host that booted once under an older version was stuck on
+  // that version's HOOK_SCRIPT forever — no upgrade could reach it. Worse, a non-atomic write
+  // interrupted mid-flight leaves a TRUNCATED md-hook.sh that existsSync() calls fine, and
+  // claude BLOCKS on that hook synchronously on every fire.
+  const dir = mkdtemp2(`${tmpdir2()}/ocp-hook-`);
+  mkdir2(dir, { recursive: true });
+  writeFile2(`${dir}/md-hook.sh`, "#!/bin/sh\n# stale, truncated leftov", { mode: 0o700 });
+  writeFile2(`${dir}/settings.json`, "{ TRUNCATED", { mode: 0o600 });
+
+  const settings = prepareStreamHook(dir);
+
+  assert.equal(readFile2(`${dir}/md-hook.sh`, "utf8"), HOOK_SCRIPT,
+    "the stale script must be replaced with the current one, not left because it existed");
+  assert.deepEqual(JSON.parse(readFile2(settings, "utf8")), buildStreamSettings(`${dir}/md-hook.sh`),
+    "…and so must the stale settings file");
+});
+
+test("stream: hook script is a write-and-exit sh script and tolerates a missing sink var", () => {
+  // forceSyncExecution: claude BLOCKS on this hook, so it must do no work inline.
+  assert.ok(HOOK_SCRIPT.startsWith("#!/bin/sh"));
+  assert.ok(HOOK_SCRIPT.includes('[ -n "$OCP_TUI_STREAM_FILE" ] || exec cat >/dev/null'),
+    "no sink configured => swallow stdin and exit 0; never fail, never block claude");
+  assert.ok(!/curl|node |python/.test(HOOK_SCRIPT), "no interpreter/network work in a blocking hook");
+});
+
+test("stream: settings registers exactly one MessageDisplay command hook (static, no per-request data)", () => {
+  const s = buildStreamSettings("/x/md-hook.sh");
+  assert.deepEqual(Object.keys(s.hooks), ["MessageDisplay"]);
+  assert.equal(s.hooks.MessageDisplay[0].hooks[0].type, "command");
+  assert.equal(s.hooks.MessageDisplay[0].hooks[0].command, "/x/md-hook.sh");
+  // Warm-pool compatibility: the settings file must NOT carry a session/request-specific path.
+  assert.ok(!JSON.stringify(s).includes(".jsonl"), "sink path comes from the pane env, not the settings file");
+});
+
+test("stream: sink path is keyed by session_id (concurrent panes cannot interleave)", () => {
+  // OCP_TUI_MAX_CONCURRENT defaults to 2 — two claude panes DO run at once. A shared sink
+  // would splice request A's deltas into request B's stream.
+  const A = streamFilePath("/d", "aaaa-1111");
+  const B = streamFilePath("/d", "bbbb-2222");
+  assert.notEqual(A, B, "one sink per session-id");
+  assert.ok(A.endsWith("/aaaa-1111.jsonl"));
+});
+
+test("stream: buildTuiCmd — OFF is byte-for-byte the pre-streaming argv; ON adds only env + --settings", () => {
+  const off = buildTuiCmd("/bin/claude", "m", "SID", "/h", "cli");
+  assert.ok(!off.includes("--settings"), "no --settings when streaming is off");
+  assert.ok(!off.includes("OCP_TUI_STREAM_FILE"), "no sink env when streaming is off");
+  const on = buildTuiCmd("/bin/claude", "m", "SID", "/h", "cli", { file: "/d/SID.jsonl", settings: "/d/s.json" });
+  assert.ok(on.includes("OCP_TUI_STREAM_FILE='/d/SID.jsonl'"), "sink delivered via the pane env");
+  assert.ok(on.includes("--settings '/d/s.json'"));
+  // must not regress the MCP wall or the pinned effort (#156)
+  assert.ok(on.includes("--strict-mcp-config") && on.includes("--disallowedTools 'mcp__*'"), "MCP wall intact");
+  assert.ok(on.includes("--effort low"), "OCP_TUI_EFFORT default intact");
+  assert.ok(!on.includes(" -p ") && !on.includes("--bare"), "still a plain interactive TUI spawn");
+});
+
+test("stream: /health block is additive and exposes the divergence counter", () => {
+  const stats = { lastEntrypoint: "cli", entrypointMismatches: 0, streamTurns: 3, streamDeltas: 21, streamTopUps: 1, streamDivergences: 0 };
+  const sem = { inflight: 0, queued: 0 };
+  const b = buildTuiHealthBlock({ enabled: true, entrypointMode: "cli", maxConcurrent: 2, streamEnabled: true }, stats, sem);
+  assert.equal(b.streamEnabled, true);
+  assert.equal(b.streamTurns, 3);
+  assert.equal(b.streamDivergences, 0);
+  // existing fields unchanged (grandfathered /health consumers)
+  assert.equal(b.enabled, true);
+  assert.equal(b.entrypointMode, "cli");
+  assert.equal(b.maxConcurrent, 2);
+  // a pre-streaming tuiStats (no stream* keys) must not produce undefined/NaN
+  const legacy = buildTuiHealthBlock({ enabled: false, entrypointMode: "cli", maxConcurrent: 2 }, { lastEntrypoint: null, entrypointMismatches: 0 }, sem);
+  assert.equal(legacy.streamEnabled, false);
+  assert.equal(legacy.streamDivergences, 0);
+});
+
 // ── Cleanup ──
-runAsyncTests().then(() => {
+// Settle the async-bodied tests registered through the sync `test()` helper BEFORE summarizing —
+// otherwise their pass/fail is not reflected in the counts (see the `pendingAsync` comment above).
+// ─── TUI streaming × warm pool: the INTEGRATION seam (backlog #2 rebased onto #158) ───
+//
+// The hook is installed by bootTuiPane at BOOT, and runTuiTurn reads the sink off the PANE
+// (pane.streamFile). That indirection is the entire reason a POOLED pane streams: the pool
+// pre-boots panes long before a request exists, so anything derived at turn time would leave
+// every pool HIT silently buffered while every MISS streamed — a perf regression with no
+// failing test and no error, visible only as "streaming mysteriously does nothing in prod".
+// These three guard that seam.
+console.log("\nTUI streaming × warm pane pool integration:");
+
+import { bootTuiPane as bootPaneUnderTest, runTuiTurn as runTurnUnderTest } from "./lib/tui/session.mjs";
+import { mkdtempSync as mkdtemp2, writeFileSync as writeFile2, mkdirSync as mkdir2, readFileSync as readFile2 } from "node:fs";
+import { tmpdir as tmpdir2 } from "node:os";
+
+// Fake tmux that records the spawned pane command and always looks ready + pasted.
+function makeTmuxRecorder() {
+  const cmds = [];
+  const tmux = (args) => {
+    cmds.push(args);
+    if (args[0] === "capture-pane") {
+      // input bar present AND the prompt visibly landed → both polls pass immediately
+      return { status: 0, stdout: "[Pasted text #1 +2 lines]\n ? for shortcuts" };
+    }
+    return { status: 0, stdout: "" };
+  };
+  return { tmux, cmds, paneCmd: () => (cmds.find((a) => a[0] === "new-session") || []).slice(-1)[0] || "" };
+}
+
+// A HOME with one already-terminal transcript for `sid`, so readTuiTranscript returns at once.
+function seedTranscript(home, sid, text) {
+  const dir = `${home}/.claude/projects/x`;
+  mkdir2(dir, { recursive: true });
+  writeFile2(`${dir}/${sid}.jsonl`, JSON.stringify({
+    type: "assistant",
+    message: { role: "assistant", content: [{ type: "text", text }], stop_reason: "end_turn" },
+    turn_duration: 1234, cc_entrypoint: "cli",
+  }) + "\n");
+}
+
+test("bootTuiPane with a streamDir installs the hook AT BOOT and hands the sink back on the pane", async () => {
+  const home = mkdtemp2(`${tmpdir2()}/ocp-t-`);
+  const streamDir = mkdtemp2(`${tmpdir2()}/ocp-s-`);
+  const rec = makeTmuxRecorder();
+  const pane = await bootPaneUnderTest({
+    model: "sonnet", claudeBin: "claude", home, realHome: home,
+    cwd: `${home}/wk`, port: 3456, tmux: rec.tmux, streamDir,
+  });
+  // The pane carries its OWN sink, named from its OWN session-id — which is what a pre-booted
+  // pool pane needs, since it is minted with no knowledge of the request it will eventually serve.
+  assert.ok(pane.streamFile, "a streamDir must yield a per-pane sink on the returned pane");
+  assert.ok(pane.streamFile.includes(pane.sessionId), "the sink is keyed by the pane's own session-id");
+  const cmd = rec.paneCmd();
+  assert.ok(cmd.includes("OCP_TUI_STREAM_FILE="), "the pane's env must carry its sink path");
+  assert.ok(cmd.includes(pane.streamFile), "…and it must be THIS pane's sink, not a shared one");
+  assert.ok(cmd.includes("--settings"), "the MessageDisplay hook must be registered at spawn");
+});
+
+test("bootTuiPane WITHOUT a streamDir spawns exactly today's pane — no hook, no --settings", async () => {
+  const home = mkdtemp2(`${tmpdir2()}/ocp-t-`);
+  const rec = makeTmuxRecorder();
+  const pane = await bootPaneUnderTest({
+    model: "sonnet", claudeBin: "claude", home, realHome: home,
+    cwd: `${home}/wk`, port: 3456, tmux: rec.tmux,
+  });
+  assert.equal(pane.streamFile, null, "no streamDir → no sink (streaming is opt-in, default OFF)");
+  const cmd = rec.paneCmd();
+  assert.ok(!cmd.includes("--settings"), "the default spawn must not gain --settings");
+  assert.ok(!cmd.includes("OCP_TUI_STREAM_FILE"), "the default spawn must not gain the hook env");
+});
+
+test("REGRESSION: a WARM (pooled) pane streams — the sink comes off the pane, not the turn", async () => {
+  const home = mkdtemp2(`${tmpdir2()}/ocp-t-`);
+  const streamDir = mkdtemp2(`${tmpdir2()}/ocp-s-`);
+  const rec = makeTmuxRecorder();
+
+  // Pre-boot a pane the way the POOL does (its own session-id + sink, fixed at boot).
+  const warm = await bootPaneUnderTest({
+    model: "sonnet", claudeBin: "claude", home, realHome: home,
+    cwd: `${home}/wk`, port: 3456, tmux: rec.tmux, streamDir,
+  });
+  // Its hook has already fired twice by the time the turn's transcript goes terminal.
+  writeFile2(warm.streamFile,
+    JSON.stringify({ hook_event_name: "MessageDisplay", delta: "Hello " }) + "\n" +
+    JSON.stringify({ hook_event_name: "MessageDisplay", delta: "world" }) + "\n");
+  seedTranscript(home, warm.sessionId, "Hello world");
+
+  const seen = [];
+  const pool = { acquire: () => warm, refill: () => {}, warm: 0 };
+  const out = await runTurnUnderTest({
+    prompt: "say hello", model: "sonnet", claudeBin: "claude", home, realHome: home,
+    cwd: `${home}/wk`, port: 3456, tmux: rec.tmux, pool,
+    onDelta: (d) => seen.push(d.delta),
+    // streamDir is deliberately NOT passed: on a pool HIT runTuiTurn never cold-boots, so if it
+    // recomputed the sink from a turn-time streamDir (the pre-rebase shape) this turn would emit
+    // ZERO deltas and silently serve buffered. Reading pane.streamFile is what makes it stream.
+    streamDir: null,
+  });
+  assert.deepEqual(seen, ["Hello ", "world"], "the pooled pane's deltas must reach the client");
+  assert.equal(out.text, "Hello world", "and the transcript stays authoritative for the final text");
+});
+
+runAsyncTests().then(() => Promise.all(pendingAsync)).then(() => {
  closeDb();
  console.log(`\n=== Results: ${passed} passed, ${failed} failed ===\n`);
  process.exit(failed > 0 ? 1 : 0);