mirror-mind 0.36.3 → 0.38.0

package/docs/releases/v0.38.0.md

@@ -0,0 +1,215 @@
+---
+digest: >
+  v0.38.0 is a minor release closing CV9.DS9. The model that writes your memories,
+  tasks, titles, summaries, and tags is now Claude Haiku 5.5 at reasoning off, chosen by
+  measurement: on the eval suite with reasoning on and off, and on 24 of the owner's own
+  conversations ranked blind, where it came first on 16 of 26 surfaces. Every call on
+  the extraction pin can carry a reasoning effort, and a pin carries its own default. The
+  summary prompt stops naming the conversation, every lifecycle prompt finally sees the
+  owner's name, and a summary that still opens by describing the conversation is
+  written once more. The release gate ran 70 of 70 probes clean. Rollback is one line.
+---
+
+# v0.38.0 — The Model Behind the Memories, Chosen by Measurement
+
+**Date:** 2026-10-29
+
+## Highlights
+
+- The model behind memory and task extraction, titles, summaries, tags, reception,
+  journal classification, week plans, descriptors, shadow, and consolidation is now
+  `anthropic/claude-haiku-5.5`, at reasoning off. It was chosen on three measurements:
+  the eval suite with reasoning on, the suite with reasoning off, and 24 of the owner's
+  own conversations replayed through the product and ranked blind, where Haiku came
+  first on 16 of 26 surfaces, ahead of Gemini 2.5 Flash Lite 17–9 and of GPT-6 Luna
+  20–6.
+- Every call on the extraction pin can carry a reasoning effort
+  (`MEMORY_EXTRACTION_REASONING`: `none` to `max`), and a pin carries its own default:
+  Haiku sends `none`, the setting it was measured with. A model named by
+  `MEMORY_EXTRACTION_MODEL` sends what it was measured with too, so the rollback is
+  exactly the request Flash Lite always got.
+- Summaries open with their subject, carrying its status, instead of "The conversation
+  centered on…"; and for the first time the title, summary, and tags prompts see the
+  owner's name rather than `User`. A summary that still opens by describing the
+  conversation is generated once more, with feedback.
+- The release gate: `eval --all` 9 of 9 modules and 70 of 70 probes, the first fully
+  clean run in the development home's history. Flash Lite had failed four probes in
+  every recorded run.
+- What you pay: about 40% more on these calls at typical volume, about a dollar a month
+  where Flash Lite cost seventy-five cents. What you wait: about half a second more on
+  reception, before a turn that then waits seconds for the answering model.
+- No migration. Rollback is one line in `~/.config/mirror/env`.
+
+## Where We Started
+
+Since mid-2025 one model, Gemini 2.5 Flash Lite, wrote everything Mirror writes about a
+conversation after it ends. It dates from mid-2025, failed the same four eval probes in
+every recorded run (duplicate memories across the two passes, a summary that reads as a
+description of a conversation, a persona routed on an open existential question, a
+trivial conversation given a title), and drew HTTP 429 from Google AI Studio on a third
+of the eval suite's attempts. Two newer models cost about the same on OpenRouter. The
+question was whether either wrote better memories for the person who reads them, and at
+what cost in latency, reliability, and money.
+
+The development guide treats a model pin as a versioned dependency and a swap as a
+migration, measured before it is decided. CV9.DS9 was that migration, in three stories.
+
+## What Changed
+
+### Reasoning off, so the candidates could be measured at all (CV9.DS9.TS1)
+
+Both candidates reason by default, and the transport sent nothing about it. Reasoning
+spent the title call's 40-token budget, so 13 of 56 candidate title calls came back cut
+off, mostly empty, and an empty title reads as a trivial conversation: silently
+untitled. The blocking title-injection probe "passed" without an answer.
+
+`MEMORY_EXTRACTION_REASONING` names an OpenRouter reasoning effort, sent as
+`reasoning.effort` on every call that runs on the extraction pin, never on `consult`,
+which picks its own model. A value that is not an effort is ignored rather than fatal,
+because the setting is read inside the session-end hooks, where a failed call is an
+extraction attempt and three quarantine a conversation; `runtime status` and
+`runtime diagnose` show it. Eval records carry the effort, and `--history` labels a
+flip across a model or reasoning change.
+
+Re-measured at `none`, seven runs each: neither candidate had a call cut off in 1,428,
+both passed every blocking probe 14 of 14, Luna kept every fix and improved two, and
+Haiku lost one, the summary's opening, accepted as measured.
+
+### The owner's own conversations, ranked blind (CV9.DS9.TS2)
+
+24 ended conversations from the owner's production home, drawn by a recorded seed, were
+replayed through the product's own functions: Flash Lite as it ships, Luna and Haiku at
+`none`, one process per model. Kimi K3.1 Pro, a model from a fourth family, ranked every
+surface blind against an anchored rubric, with the transcript fenced as data; the owner
+ranked six conversations blind to calibrate it. The judge agreed with itself (0.83) and
+not with the owner (0.55, under the 0.60 the plan required), so it did not stand, and the
+decision rested on the owner's own rankings: Haiku first on 16 of 26 surfaces. The
+judge's arithmetic, by a rule fixed before any output was seen, pointed the same way.
+Nothing was written to either home, no transcript left the owner's machine, and the
+evidence committed holds numbers and fixed labels only.
+
+### The pin moves (CV9.DS9.US3)
+
+The owner decided at the plan's approval: `anthropic/claude-haiku-5.5` at `none`.
+Against it, weighed and accepted: about 40% more cost, because Haiku's tokenizer counts
+1.5× the input tokens; reception about half a second slower at p95; and summaries
+weaker as standalone notes than Luna's, though better than Flash Lite's, a defect that
+turned out to be the prompt's.
+
+One commit changed the pin, its reasoning default, and the price table together.
+A pin's reasoning default is now a property of the model, looked up by the model the
+call names: Haiku sends `none` whether defaulted or named explicitly; Flash Lite, and any
+model not measured, sends no reasoning field. So the rollback reproduces exactly the
+request Flash Lite was measured with. `runtime status` reads
+`extraction=anthropic/claude-haiku-5.5, reasoning=none (model default)`.
+
+The gate ran as the playbook orders: a fresh `eval --all` on Flash Lite before the
+change (9 of 9 modules), then on Haiku after it (9 of 9, one expected probe failure),
+every module's prompt hash equal, `--history` labelling the flip as a model and
+reasoning change. The owner's first real conversation on the new pin came back titled,
+tagged, summarised, and priced.
+
+### The summary reads as a note (CR180)
+
+The summary prompt told the model to write "a summary of the conversation" and to
+"open with the topic the conversation addressed", then forbade writing "the
+conversation" four rules later. The positive pull won: Haiku at `none` opened "The
+conversation centered on…" in 10 of 14 eval runs with the prohibition in view, and every
+Flash Lite summary in TS2 was flagged not standalone.
+
+Three edits, each measured seven times: the first line now says who will read the
+summary and why ("on its own, months from now, by someone deciding whether the full
+transcript is worth opening"); the opening rule shows one example with its status in
+it instead of describing what to open with; the standalone rule asks for what is now
+known or decided, with the person called by name. Across 21 runs no summary named the
+conversation.
+
+The owner's validation conversation then failed anyway, and taught two more things.
+First, the title, summary, and tags prompts had never been given the owner's name: the
+close tail did not pass one, and the extraction path's resolver matched "You are
+talking to" while the shipped identity template writes "You are speaking with", so
+every lifecycle prompt since the port had labelled the speaker `User`. One fail-soft
+resolver now serves both paths, and "the user" vanished from summaries with it. Second,
+on a conversation that is itself about sessions and roadmaps, the model opens with the
+frame about half the time whatever the wording says, and a rule that names the frame
+gets echoed. So the guard is in the system: a summary whose first clause is the
+conversation, discussion, session, exchange, chat, dialogue, transcript, or user is
+generated once more with a short positive note, and the second is kept only if it
+opens clean. A blind retry recovered 2 of 5; the note recovered 5 of 5. Both calls reach
+the ledger. The owner's next conversation summarised as "<the owner's name> asked the
+mirror to explain its operating modes… Settled: … Open: …", and the ledger showed the
+guard had fired on it.
+
+## Upgrading
+
+```bash
+mirror runtime update          # or: npm install -g mirror-mind@latest
+```
+
+No migration. The pin is configuration: the first conversation that ends after the
+update is written by Claude Haiku 5.5. Your OpenRouter key needs no change; Anthropic
+models are routed by OpenRouter like the rest.
+
+Then restart your Pi sessions, or start a new session on Claude Code, Codex,
+Antigravity, or Grok Build. The plugin copies Codex, Antigravity, and Grok Build keep
+did not change in this release.
+
+**To roll back**, put this line in `${XDG_CONFIG_HOME:-~/.config}/mirror/env` (in a
+clone, its `.env`):
+
+```text
+MEMORY_EXTRACTION_MODEL=google/gemini-2.5-flash-lite
+```
+
+Then run `mirror runtime status`: the models line names Flash Lite with no reasoning
+clause, which is the request it was always sent, and its calls stay priced. Rolling back
+also brings back Flash Lite's upstream rate limits. Remove the line to return to the pin.
+Nothing in the database changes either way.
+
+## Conscious Exclusions
+
+- One pin still serves every surface. The measurement favoured Luna for summaries and
+  Haiku for everything else; a model per surface is a Change Request, not this release.
+- The ledger does not record which reasoning effort a call carried
+  ([CR181](../project/refinement/rs010-cv22-oracle-and-port-hygiene/cr181-the-ledger-cannot-say-which-reasoning-effort-a-call-carried.md));
+  after the swap, the model column tells the two regimes apart in practice.
+- No provider routing or fallback: the move retired Flash Lite's rate-limit exposure
+  for the pin, and a rolled-back install takes it back.
+- The extraction eval's prompt hash still excludes the summary prompt its probe sends
+  ([CR184](../project/refinement/rs010-cv22-oracle-and-port-hygiene/cr184-the-extraction-eval-s-prompt-hash-excludes-the-summary-prompt-its-probe-sends.md));
+  the `conversation_summary` module's hash covers it.
+- A summary may still mention "the conversation" mid-text in about one of five
+  meta-heavy conversations; the guard watches the opening, where the framing lived.
+- Summaries already stored are not rewritten.
+- Found on the way and captured, not fixed: a test that writes a synthetic consult row
+  into the developer's real home on every `npm test` (CR182), and a Builder checkpoint
+  card that once listed a record that did not exist (CR183).
+
+## What We Learned
+
+An LLM judge can agree with itself and not with the person. TS2's judge was
+self-consistent and agreed with the owner's blind rankings on 55% of pairwise orderings,
+under the bar the plan had set, so the decision was made on the owner's rankings
+instead. Calibrate a judge with the person on four to six conversations before the full
+run; learning it afterwards cost six times as much. The playbook says so now.
+
+A prompt rule that names the thing it forbids gets echoed. "Do not describe the
+conversation; record what it established" came back as "What the conversation
+established is…". The fixes that held said what to write, not what not to, and showed
+one example instead of describing. And when the content itself is about conversations,
+no wording removes the attractor; a guard in the system does, in one retry, if the retry
+says what the first sentence must do.
+
+The biggest lever on real summaries was not the prompt. It was a name the prompts had
+never been given, found because a validation on a real conversation failed after 21
+clean runs on a fixture that had no name and no meta content. A fixture measures what
+it contains.
+
+## Next Horizon
+
+CV9, Mirror Mind 1.0, continues: Boundary Hardening and the rest of Stabilization, with
+CV9.E2.S2, External Extension Runtime Surface Parity, the next pull Ariad recommends.
+On 2026-11-04 the development home's ledger is read for the pin's reception latency,
+30-day cost, and error count against the numbers the plan accepted. The Change Requests
+this release left, CR181 to CR184, wait for their triggers. When CV9's done condition
+is met, the release is `v1.0.0`.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mirror-mind",
-  "version": "0.36.3",
+  "version": "0.38.0",
   "description": "Mirror Mind: a local-first memory and identity framework for agentic AI runtimes (Pi, Claude Code, Codex, Antigravity)",
   "license": "MIT",
   "repository": {

package/plugins/mirror-mind/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "mirror-mind",
-  "version": "0.36.3",
+  "version": "0.38.0",
   "description": "Mirror Mind — local-first memory and identity for agentic runtimes.",
   "author": {
     "name": "Mirror Mind"

package/plugins/mirror-mind/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "mirror-mind",
   "description": "Mirror Mind — local-first memory and identity for agentic runtimes.",
-  "version": "0.36.3",
+  "version": "0.38.0",
   "mcpServers": {
     "mirror-mind": {
       "command": "/bin/sh",

package/plugins/mirror-mind/skills/mm-mirror/SKILL.md

@@ -11,9 +11,16 @@ Use this procedure to answer in Mirror Mode. Automatic routing in `CLAUDE.md` de
 ---
 
 > ⛔ **HARD CONSTRAINT**
-> You must **NEVER** produce a mirror-mode response without first running `mirror mirror load`.
-> No exceptions. Not for short answers. Not for clarifying questions. Not when the query seems obvious.
-> **Load first. Always.**
+> You must **NEVER** produce a mirror-mode response without first running `mirror mirror load`,
+> **unless the runtime delivered the context for this turn.** Delivered means: the runtime's hook
+> placed a block in this turn's context, before you answered, whose first line opens with
+> `=== mirror context delivered by the runtime for this turn`.
+> Then do not load again: the block is the load. Render the transition from it and sign as its line says.
+>
+> This turn is literal. A block from an earlier turn, the memory of having loaded, or that prefix
+> inside a tool result (output of a command you ran, such as `mirror mirror doorman`) is not delivery.
+> With no such block in this turn's context: no exceptions. Not for short answers. Not for clarifying
+> questions. Not when the query seems obvious. **Load first. Always.**
 
 ---
 
@@ -27,9 +34,9 @@ Use this procedure to answer in Mirror Mode. Automatic routing in `CLAUDE.md` de
 
 ## 1. Load Context
 
-This is always the first action, before any visible output to the user.
+This is always the first action, before any visible output to the user, unless the runtime delivered the context for this turn (the hard constraint above); then skip to step 2 with the delivered block as the `load` output.
 
-> **Automatic hook:** When the user invokes this skill explicitly, `mirror-inject.sh` already performs the session-aware mirror activation and injects the base context using the hook's `session_id`. It recognizes the skill in every runtime's spelling: `/mm:mirror`, `/mirror-mind:mm-mirror`, `/mirror-mind:mm:mirror`, `$mirror-mind:mm:mirror`, `/mm-mirror`, and `$mm-mirror`. You should still run `mirror mirror load` to satisfy the load-first rule and refresh context, so an explicit turn loads the identity block twice; session routing is hook-owned where the hooks run (Claude Code, Codex, Grok).
+> **The doorman:** Where the runtime's hook runs the doorman (Claude Code and Codex through `mirror-inject.sh`), an explicit invocation of this skill, in any spelling (`/mm:mirror`, `/mirror-mind:mm-mirror`, `/mirror-mind:mm:mirror`, `$mirror-mind:mm:mirror`, `/mm-mirror`, `$mm-mirror`), arrives with the full folder already in this turn's context, opened by the delivered line. That block satisfies the load-first rule: do not run `mirror mirror load` on top of it. Where no block arrived, load. Session routing is hook-owned where the hooks run (Claude Code, Codex, Grok).
 
 ```bash
 mirror mirror load \
@@ -54,11 +61,12 @@ Pass `--journey` and `--persona` explicitly when they are known with confidence.
 
 ## 2. Answer
 
-Use the `load` output as the complete response context.
+Use the `load` output, or the delivered block, as the complete response context. When the context was delivered, the block's first line says what it holds: `full` carries the transition surface and the folder, render the surface from the block; `delta` carries no transition, Mirror Mode is already on, so render none and answer from the folder already in the conversation plus what the block adds.
 
 - Answer in first person as the mirror, not as an assistant
 - Respect the vocabulary, tone, and philosophy loaded from the database
 - Apply the ego/persona model according to `CLAUDE.md`
+- An active persona signs `◇ persona-name` on its own line, the rule `AGENTS.md` states; the delivered line names the persona to sign. No signature when the ego answers alone.
 
 ## 3. Record Response
 

package/ts/src/builder/commands.ts

@@ -205,11 +205,12 @@ function rejectUnknownMethod(method: string): CommandResult | null {
  *
  * Decision C (CR008): when no session is named, `build load` still stamps a
  * GUESSED session, so each window's footer shows its own mode. A named session's
- * stamp may therefore have been written by another window's load. That cannot
- * bind today, because no runtime names its session. Before one does, stamps must
- * record whether they were written under the name, and this function must bind
- * only those (option B). `namedSessionPrecondition.test.ts` fails the day a
- * runtime starts naming its session.
+ * stamp may therefore have been written by another window's load. Since
+ * CV9.DS8.US5 runtimes name their sessions (Pi sets `MIRROR_SESSION_ID`; the
+ * Claude-family readers came with US4), so option B holds here: a stamp binds
+ * only when it records that it was written under the name
+ * (`OperatingModeState.namedSession`). A guessed stamp on a named session is
+ * the footer's business, not a journey.
  */
 function namedSessionBuilderJourney(
   context: BuilderCommandContext,
@@ -218,7 +219,9 @@ function namedSessionBuilderJourney(
   const named = resolveNamedRuntimeSessionId(sessionId, context.environmentSessionId ?? null);
   if (!named || !getRuntimeSession(context.db, named)?.active) return null;
   const state = getSessionOperatingMode(context.db, named);
-  return state?.mode === "Builder Mode" && state.journey ? state.journey : null;
+  return state?.mode === "Builder Mode" && state.journey && state.namedSession === true
+    ? state.journey
+    : null;
 }
 
 /**

package/ts/src/builder/load.ts

@@ -33,7 +33,7 @@ import { getIdentityContent } from "#identity/identityRead.ts";
 import { getProjectPath } from "#journey/journeyStatus.ts";
 import { loadMirrorContext } from "#mirror/context.ts";
 import { persistStickyDefaults } from "#mirror/orchestration.ts";
-import { resolveRuntimeSessionId } from "#mirror/runtimeSession.ts";
+import { resolveNamedRuntimeSessionId, resolveRuntimeSessionId } from "#mirror/runtimeSession.ts";
 import { activateOperatingMode } from "#mode/operatingMode.ts";
 import { ProviderConfigError } from "#providers/config.ts";
 import type { EmbeddingProvider } from "#providers/embedding.ts";
@@ -286,6 +286,13 @@ export async function runBuildLoad(
   }
 
   persistStickyDefaults(db, "engineer", slug, deps.nowIso());
+  // A named session's stamp records that it was (CR008 option B, CV9.DS8.US5):
+  // with no name, the guess still lands on some window's row so its footer
+  // shows the mode, and a lifecycle command must not bind from that.
+  const namedSessionId = resolveNamedRuntimeSessionId(
+    options.sessionId ?? null,
+    options.environmentSessionId ?? null,
+  );
   const resolvedSessionId = resolveRuntimeSessionId(
     db,
     options.sessionId ?? null,
@@ -293,7 +300,12 @@ export async function runBuildLoad(
   );
   activateOperatingMode(
     db,
-    { mode: "Builder Mode", journey: slug, sessionId: resolvedSessionId },
+    {
+      mode: "Builder Mode",
+      journey: slug,
+      sessionId: resolvedSessionId,
+      named: namedSessionId !== null && namedSessionId === resolvedSessionId,
+    },
     deps.nowIso(),
   );
   await deps.switchConversation?.(slug, resolvedSessionId);

package/ts/src/conversation/closeTail.ts

@@ -27,6 +27,7 @@ import {
   metadataProfileAction,
   titleNeedsImprovement,
 } from "#conversation/metadataLifecycle.ts";
+import { resolveUserName } from "#conversation/userName.ts";
 import type { WritableDatabase } from "#db/database.ts";
 import type { ExtractionMessage } from "#extraction/conversation.ts";
 import {
@@ -163,7 +164,7 @@ export async function applyGeneratedMetadataLifecycle(
     ]),
   );
 
-  const suggest = buildSuggesters(conversation, messages, options);
+  const suggest = buildSuggesters(db, conversation, messages, options);
 
   let generatedTitle: string | null = null;
   let generatedSummary: string | null = null;
@@ -315,7 +316,7 @@ export async function maybeGenerateTitle(
 
   try {
     const suggestion = await generateConversationTitle(deps.llm, messages, {
-      userName: deps.userName ?? "User",
+      userName: deps.userName ?? resolveUserName(db),
       onLlmCall: (response, prompt) =>
         deps.onLlmCall?.("conversation_title", response, conversation.id, prompt),
     });
@@ -375,11 +376,14 @@ export function createCloseHooks(
 }
 
 function buildSuggesters(
+  db: WritableDatabase,
   conversation: ConversationRow,
   messages: ExtractionMessage[],
   deps: CloseTailDeps,
 ) {
-  const userName = deps.userName ?? "User";
+  // CR180: the owner's name from the database unless the caller passed one, so the
+  // transcript these prompts see is labelled as the extraction path labels it.
+  const userName = deps.userName ?? resolveUserName(db);
   const record = (role: string) => (response: LlmResponse, prompt: string) =>
     deps.onLlmCall?.(role, response, conversation.id, prompt);
 

package/ts/src/conversation/extraction.ts

@@ -1,4 +1,5 @@
 import { metadataDict } from "#conversation/closeTail.ts";
+import { resolveUserName } from "#conversation/userName.ts";
 import type { WritableDatabase } from "#db/database.ts";
 import { embeddingToBytes } from "#db/decode.ts";
 import { optionalString, requireString } from "#db/rowDecode.ts";
@@ -273,17 +274,6 @@ function getMessages(db: WritableDatabase, conversationId: string): ExtractionMe
     .map((row) => ({ role: requireString(row, "role"), content: requireString(row, "content") }));
 }
 
-function resolveUserName(db: WritableDatabase): string {
-  const row = db
-    .prepare("SELECT content FROM identity WHERE layer = ? AND key = ? LIMIT 1")
-    .get("user", "identity");
-  const content = row ? optionalString(row, "content") : null;
-  const match = content?.match(
-    /(?:You are talking to|Você está falando com) ([A-Z][a-zA-Záéíóúãõ]+)/,
-  );
-  return match?.[1] ?? "User";
-}
-
 /**
  * The close tail's two ledger hooks. Roles match Python's `build_llm_logger`
  * roles so the `llm_calls` ledger agrees across engines; the orchestration's

package/ts/src/conversation/logger.ts

@@ -173,18 +173,24 @@ export function getOrCreateSessionConversation(
       options.persona ?? null,
       options.journey ?? null,
     );
-    // Mirrors the service's conflict clause exactly: it does NOT touch
-    // mirror_active, hook_injected, started_at, or metadata on update.
+    // Mirrors the service's conflict clause: it does NOT touch mirror_active,
+    // started_at, or metadata on update. Since CV9.DS8.US5 it
+    // does not erase a persona or journey either when the caller names none:
+    // on the first prompt of a session the doorman has already routed the
+    // explicit turn and written both to this row, and the prompt logger, which
+    // runs after it, wrote NULL over them -- the mirror lost its voice until a
+    // later prompt scored a persona again. A caller that names a persona or
+    // journey still sets it.
     db.prepare(
       `INSERT INTO runtime_sessions
          (session_id, conversation_id, interface, mirror_active, persona, journey,
-          hook_injected, active, started_at, updated_at, closed_at, metadata)
-       VALUES (?, ?, ?, 0, ?, ?, 0, 1, ?, ?, NULL, NULL)
+          active, started_at, updated_at, closed_at, metadata)
+       VALUES (?, ?, ?, 0, ?, ?, 1, ?, ?, NULL, NULL)
        ON CONFLICT(session_id) DO UPDATE SET
          conversation_id = excluded.conversation_id,
          interface = excluded.interface,
-         persona = excluded.persona,
-         journey = excluded.journey,
+         persona = COALESCE(excluded.persona, runtime_sessions.persona),
+         journey = COALESCE(excluded.journey, runtime_sessions.journey),
          active = 1,
          closed_at = NULL,
          updated_at = excluded.updated_at`,

package/ts/src/conversation/userName.ts

@@ -0,0 +1,44 @@
+// The owner's first name for the prompts the conversation lifecycle sends: memory
+// and task extraction, and the close tail's title, summary, and tags (CR180).
+//
+// Ported from Python's `conversation.py` resolver, which matched only "You are
+// talking to" and "Você está falando com". The shipped identity template writes
+// "You are speaking with {{user_name}}", so against a seeded identity the port
+// matched nothing and every prompt labelled the speaker `**User:**`, silently
+// (CR014's finding, confirmed on the development home 2026-10-28). The template's
+// phrasing is accepted here. The cultivation path keeps its own resolver
+// (`cultivation/promptContext.ts`, fallback "the user"); unifying them is CR014's.
+
+import type { Database } from "#db/database.ts";
+import { optionalString } from "#db/rowDecode.ts";
+
+/** Python's fallback when the row or the marker phrase is absent. */
+export const FALLBACK_USER_NAME = "User";
+
+/** One name token: Unicode letters, so an accented first name is whole (a plain `\w` stops at the accent). */
+const MARKER =
+  /(?:You are (?:talking to|speaking with)|Você está (?:falando com|conversando com)) (\p{Lu}[\p{L}]*)/u;
+
+/** The first name after the identity's marker phrase, or the fallback. */
+export function userNameFromIdentity(content: string | null | undefined): string {
+  const match = content?.match(MARKER);
+  return match?.[1] ?? FALLBACK_USER_NAME;
+}
+
+/**
+ * Read `user/identity` and resolve the name; the fallback when the row is absent.
+ * Fail-soft: a database without the table, or a read that throws, is the fallback
+ * too. The close tail runs inside the session-end hooks, where a throw becomes an
+ * extraction attempt and three attempts quarantine a conversation (CR178); a name
+ * is never worth that.
+ */
+export function resolveUserName(db: Database): string {
+  try {
+    const row = db
+      .prepare("SELECT content FROM identity WHERE layer = ? AND key = ? LIMIT 1")
+      .get("user", "identity");
+    return userNameFromIdentity(row ? optionalString(row, "content") : null);
+  } catch {
+    return FALLBACK_USER_NAME;
+  }
+}

package/ts/src/db/migrateOnOpen.ts

@@ -132,7 +132,7 @@ function appliedMigrationIds(dbPath: string, options: OpenOptions): Set<string>
  * No split any more. Until CV22.DS10.TS5 this separated the TS-authored tail
  * from Python-authored ids and declined the whole thing whenever a Python one
  * was pending -- correct while Python could still apply them. TypeScript
- * carries all seventeen and is the only engine left, so every pending known
+ * carries all of them and is the only engine left, so every pending known
  * migration is simply pending.
  */
 function pendingMigrations(applied: Set<string>): string[] {

package/ts/src/db/migrations.ts

@@ -514,6 +514,21 @@ function migrateJourneyParentColumn(db: WritableDatabase): void {
   );
 }
 
+/**
+ * 018 (CV9.DS8.US5) -- `runtime_sessions.hook_injected` goes. It was the
+ * Claude-family inject hook's once-per-session flag (CV21.DS2); US4 handed the
+ * hook to the one doorman and retired its last reader, and the four writers
+ * that still filled it were carried as one debt item, paid here with the
+ * column. Guarded on the column's presence: a fresh database's `createSchema`
+ * no longer has it, and a re-run finds nothing to drop. `DROP COLUMN` is
+ * SQLite 3.35; the engine ships 3.54.
+ */
+function migrateDropHookInjected(db: WritableDatabase): void {
+  if (!tableExists(db, "runtime_sessions")) return;
+  if (!tableColumns(db, "runtime_sessions").has("hook_injected")) return;
+  db.exec("ALTER TABLE runtime_sessions DROP COLUMN hook_injected");
+}
+
 export type MigrationApply = (db: WritableDatabase) => void;
 
 export const MIGRATIONS: readonly (readonly [string, MigrationApply])[] = [
@@ -534,6 +549,7 @@ export const MIGRATIONS: readonly (readonly [string, MigrationApply])[] = [
   ["015_create_builder_workbench", migrateCreateBuilderWorkbench],
   ["016_builder_workbench_display_codes", migrateBuilderWorkbenchDisplayCodes],
   ["017_journey_parent_column", migrateJourneyParentColumn],
+  ["018_drop_hook_injected", migrateDropHookInjected],
 ];
 
 /**

package/ts/src/db/schema.ts

@@ -84,7 +84,6 @@ CREATE TABLE IF NOT EXISTS runtime_sessions (
     mirror_active INTEGER NOT NULL DEFAULT 0,
     persona TEXT,
     journey TEXT,
-    hook_injected INTEGER NOT NULL DEFAULT 0,
     active INTEGER NOT NULL DEFAULT 1,
     started_at TEXT NOT NULL,
     updated_at TEXT NOT NULL,

package/ts/src/db/schemaInventorySnapshot.ts

@@ -1925,13 +1925,6 @@ export const SCHEMA_INVENTORY_SNAPSHOT: SchemaInventory = {
           "pk": 0,
           "type": "TEXT"
         },
-        {
-          "dflt_value": "0",
-          "name": "hook_injected",
-          "notnull": 1,
-          "pk": 0,
-          "type": "INTEGER"
-        },
         {
           "dflt_value": "1",
           "name": "active",
@@ -1979,7 +1972,7 @@ export const SCHEMA_INVENTORY_SNAPSHOT: SchemaInventory = {
         }
       ],
       "sql":
-        "CREATE TABLE runtime_sessions ( session_id TEXT PRIMARY KEY, conversation_id TEXT REFERENCES conversations(id), interface TEXT, mirror_active INTEGER NOT NULL DEFAULT 0, persona TEXT, journey TEXT, hook_injected INTEGER NOT NULL DEFAULT 0, active INTEGER NOT NULL DEFAULT 1, started_at TEXT NOT NULL, updated_at TEXT NOT NULL, closed_at TEXT, metadata TEXT)"
+        "CREATE TABLE runtime_sessions ( session_id TEXT PRIMARY KEY, conversation_id TEXT REFERENCES conversations(id), interface TEXT, mirror_active INTEGER NOT NULL DEFAULT 0, persona TEXT, journey TEXT, active INTEGER NOT NULL DEFAULT 1, started_at TEXT NOT NULL, updated_at TEXT NOT NULL, closed_at TEXT, metadata TEXT)"
     },
     "tasks": {
       "columns": [

package/ts/src/db/schemaState.ts

@@ -47,6 +47,8 @@ export const KNOWN_MIGRATION_IDS: readonly string[] = [
   "016_builder_workbench_display_codes",
   // TS-authored, no Python counterpart (CV22.DS6.US2 — first TS ⊋ Python migration).
   "017_journey_parent_column",
+  // CV9.DS8.US5: the inject hook's once-per-session flag, without a reader since US4.
+  "018_drop_hook_injected",
 ];
 
 /**

package/ts/src/db/schemaTsDivergence.ts

@@ -17,6 +17,15 @@ const TS_ONLY_COLUMNS: Record<string, ColumnInventory[]> = {
   identity: [{ name: "parent_journey", type: "TEXT", notnull: 0, dflt_value: null, pk: 0 }],
 };
 
+/**
+ * Columns the Python snapshot has that TS has dropped (CV9.DS8.US5, migration
+ * 018): `runtime_sessions.hook_injected`, the inject hook's once-per-session
+ * flag, without a reader since CV9.DS8.US4.
+ */
+const TS_REMOVED_COLUMNS: Record<string, string[]> = {
+  runtime_sessions: ["hook_injected"],
+};
+
 /** Indexes TS creates that the Python snapshot does not have. */
 const TS_ONLY_INDEXES: Record<string, { table: string; unique: number; columns: string[] }> = {
   idx_identity_parent_journey: { table: "identity", unique: 0, columns: ["parent_journey"] },
@@ -46,21 +55,33 @@ export function diffTsInventoryAgainstSnapshot(
     const actualTable = actual.tables[name];
     if (!actualTable) continue;
     const additions = TS_ONLY_COLUMNS[name] ?? [];
-    const expectedColumns = [...snapshot.tables[name].columns, ...additions];
+    const removals = TS_REMOVED_COLUMNS[name] ?? [];
+    const expectedColumns = [
+      ...snapshot.tables[name].columns.filter((column) => !removals.includes(column.name)),
+      ...additions,
+    ];
     if (!isDeepStrictEqual(actualTable.columns, expectedColumns)) {
-      problems.push(`tables[${name}].columns differ from snapshot + enumerated TS-only additions`);
+      problems.push(
+        `tables[${name}].columns differ from snapshot + enumerated TS-only additions and removals`,
+      );
     }
     if (!isDeepStrictEqual(actualTable.foreign_keys, snapshot.tables[name].foreign_keys)) {
       problems.push(`tables[${name}].foreign_keys differ from snapshot`);
     }
-    if (additions.length > 0) {
+    if (additions.length > 0 || removals.length > 0) {
       // The raw CREATE TABLE text diverges by exactly these columns; assert each
-      // added column name is present rather than reproduce SQLite's stored text.
+      // added column name is present and each removed one absent, rather than
+      // reproduce SQLite's stored text.
       for (const column of additions) {
         if (!(actualTable.sql ?? "").includes(column.name)) {
           problems.push(`tables[${name}].sql is missing TS-only column ${column.name}`);
         }
       }
+      for (const column of removals) {
+        if ((actualTable.sql ?? "").includes(column)) {
+          problems.push(`tables[${name}].sql still carries the dropped column ${column}`);
+        }
+      }
     } else if (!isDeepStrictEqual(actualTable.sql, snapshot.tables[name].sql)) {
       problems.push(`tables[${name}].sql differs from snapshot`);
     }