chore: bump version to 12.2.0

Merge pull request #2052 from thedotmack/thedotmack/worktree-remap
feat(worktree): scope per worktree, cwd backfill, and merged-worktree adoption
2026-04-17 20:14:22 -07:00 · 2026-04-17 20:10:54 -07:00 · 2026-04-16 22:01:21 -07:00 · 2026-04-16 21:31:30 -07:00 · 2026-04-16 20:05:37 -07:00 · 2026-04-16 20:03:27 -07:00
67 changed files with 2304 additions and 1894 deletions
@@ -1,136 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Oct 25, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #2374 | 2:55 PM | ✅ | Marketplace metadata version synchronized to 4.2.11 | ~157 |
-
-### Oct 27, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #2757 | 1:23 AM | 🟣 | Released v4.3.3 with Configurable Session Display and First-Time Setup UX | ~391 |
-
-### Nov 4, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #3706 | 9:47 PM | ✅ | Marketplace Plugin Version Synchronized to 5.0.2 | ~162 |
-| #3655 | 3:43 PM | ✅ | Version bumped to 5.0.1 across project | ~354 |
-
-### Nov 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4068 | 10:58 PM | ✅ | Committed v5.1.0 release with comprehensive release notes | ~486 |
-| #4066 | 10:57 PM | ✅ | Updated marketplace.json version to 5.1.0 | ~192 |
-| #3739 | 2:24 PM | ✅ | Updated version to 5.0.3 across project manifests | ~322 |
-
-### Nov 6, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4099 | 1:13 PM | 🟣 | Theme Toggle for Light/Dark Mode | ~253 |
-| #4096 | " | ✅ | Marketplace Metadata Version Sync | ~179 |
-| #4092 | 1:12 PM | 🔵 | Marketplace Configuration for Claude-Mem Plugin | ~194 |
-| #4078 | 12:50 PM | 🔴 | Fixed PM2 ENOENT error on Windows systems | ~286 |
-| #4075 | 12:49 PM | ✅ | Marketplace plugin version synchronized to 5.1.1 | ~189 |
-
-### Nov 7, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4612 | 6:33 PM | ✅ | Version Bumped to 5.2.0 Across All Package Metadata | ~359 |
-| #4598 | 6:31 PM | ✅ | PR #69 Merged: cleanup/worker Branch Integration | ~469 |
-| #4298 | 11:54 AM | 🔴 | Fixed PostToolUse Hook Schema Compliance | ~310 |
-| #4295 | 11:53 AM | ✅ | Synchronized Plugin Marketplace Version to 5.1.4 | ~188 |
-
-### Nov 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #5150 | 7:37 PM | 🟣 | Troubleshooting Skill Added to Claude-Mem Plugin | ~427 |
-| #5133 | 7:29 PM | ✅ | Version 5.2.3 Released with Build Process | ~487 |
-
-### Nov 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #5941 | 7:14 PM | ✅ | Marketplace Version Updated to 5.4.0 | ~157 |
-
-### Nov 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6341 | 1:49 PM | ✅ | Version Bumped to 5.4.1 | ~239 |
-
-### Nov 11, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6602 | 1:51 PM | ✅ | Version 5.4.5 Released to GitHub | ~279 |
-| #6601 | " | ✅ | Version Patch Bump 5.4.4 to 5.4.5 | ~233 |
-
-### Nov 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #8212 | 3:06 PM | 🔵 | Version Consistency Verification Across Multiple Configuration Files | ~238 |
-
-### Nov 25, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #14882 | 1:32 PM | 🔵 | Marketplace Configuration Defines Plugin Version and Source Directory | ~366 |
-
-### Nov 30, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #18064 | 10:52 PM | ✅ | Bumped version to 6.3.7 in marketplace.json | ~179 |
-| #18060 | 10:51 PM | 🔵 | Read marketplace.json plugin manifest | ~190 |
-
-### Dec 1, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #18428 | 3:33 PM | 🔵 | Version Conflict in Marketplace Configuration | ~191 |
-
-### Dec 4, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20049 | 3:23 PM | ✅ | Updated marketplace.json version to 6.5.2 | ~203 |
-
-### Dec 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22559 | 1:08 AM | ✅ | Version 7.0.3 committed to repository | ~261 |
-| #22551 | 1:07 AM | ✅ | Marketplace metadata updated to version 7.0.3 | ~179 |
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23440 | 2:25 PM | ✅ | Marketplace Configuration Updated to 7.0.8 | ~188 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26799 | 11:39 PM | ✅ | Marketplace Manifest Version Updated to 7.2.3 | ~248 |
-| #26796 | " | ✅ | Version Bumped to 7.2.3 in marketplace.json | ~259 |
-| #26792 | 11:38 PM | 🔵 | Current Version Confirmed as 7.2.2 Across All Configuration Files | ~291 |
-
-### Dec 16, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #28306 | 10:08 PM | 🔵 | Marketplace Configuration Also Shows Version 7.3.3 | ~220 |
-| #27555 | 4:48 PM | ✅ | Version bump committed to main branch | ~242 |
-| #27553 | " | ✅ | Version consistency verified across all configuration files | ~195 |
-| #27551 | 4:47 PM | ✅ | Marketplace.json version updated to 7.3.1 | ~207 |
-</claude-mem-context>
@@ -10,7 +10,7 @@
  "plugins": [
    {
      "name": "claude-mem",
-      "version": "12.1.2",
+      "version": "12.2.0",
      "source": "./plugin",
      "description": "Persistent memory system for Claude Code - context compression across sessions"
    }
@@ -1,6 +1,6 @@
 {
  "name": "claude-mem",
-  "version": "12.1.1",
+  "version": "12.2.0",
  "description": "Memory compression system for Claude Code - persist context across sessions",
  "author": {
    "name": "Alex Newman"
@@ -1,29 +0,0 @@
-# Project-Level Skills
-
-This directory contains skills **for developing and maintaining the claude-mem project itself**, not skills that are released as part of the plugin.
-
-## Distinction
-
-**Project Skills** (`.claude/skills/`):
- Used by developers working on claude-mem
- Not included in the plugin distribution
- Project-specific workflows (version bumps, release management, etc.)
- Not synced to `~/.claude/plugins/marketplaces/thedotmack/`
-
-**Plugin Skills** (`plugin/skills/`):
- Released as part of the claude-mem plugin
- Available to all users who install the plugin
- General-purpose memory search functionality
- Synced to user installations via `npm run sync-marketplace`
-
-## Skills in This Directory
-
-### version-bump
-Manages semantic versioning for the claude-mem project itself. Handles updating all three version files (package.json, marketplace.json, plugin.json), creating git tags, and GitHub releases.
-
-**Usage**: Only for claude-mem maintainers releasing new versions.
-
-## Adding New Skills
-
-**For claude-mem development** → Add to `.claude/skills/`
-**For end users** → Add to `plugin/skills/` (gets distributed with plugin)
@@ -1,6 +1,6 @@
 {
  "name": "claude-mem",
-  "version": "12.1.1",
+  "version": "12.2.0",
  "description": "Memory compression system for Claude Code - persist context across sessions",
  "author": {
    "name": "Alex Newman",
@@ -0,0 +1,570 @@
+# Merged-Worktree Adoption
+
+**Goal**: When a worktree's branch is merged into its parent, the worktree's observations become part of the parent project's observation list — without data movement, destructive schema changes, or lost provenance.
+
+**Approach**: Add a nullable `merged_into_project` column to observations and session_summaries, extend query predicates with `OR merged_into_project = :parent`, propagate the same metadata to Chroma embeddings for semantic-search consistency, detect merges via git (authoritative), run adoption automatically on worker startup, and offer a CLI escape hatch for squash-merges.
+
+**Key design decisions**:
+- `observations.project` is **immutable provenance** — never overwritten.
+- Merged-status is a **virtual pointer**, not a data move.
+- **Chroma metadata stays in lockstep with SQLite** (full consistent sync, not lazy SQL expansion). Single source of truth per row.
+- Detection is **git-authoritative** (`git worktree list --porcelain` + `git branch --merged`), with a manual CLI override for squash-merges.
+
+---
+
+## Phase 0 — Documentation Discovery (COMPLETE)
+
+Findings consolidated from three parallel discovery subagents. The following are the ONLY APIs/patterns to copy from. Do not invent alternatives.
+
+### Allowed APIs (copy from these locations)
+
+| Need | File | Lines | What to copy |
+|---|---|---|---|
+| Migration idempotency via marker file | `src/services/infrastructure/ProcessManager.ts` | 680–830 | `runOneTimeCwdRemap` structure, marker file pattern `.cwd-remap-applied-v1` |
+| Worker startup wiring | `src/services/worker-service.ts` | 363–365 | Call site inside `initializeBackground()`, invoked before `dbManager.initialize()` |
+| `ALTER TABLE ADD COLUMN` idempotency | `src/services/sqlite/migrations/runner.ts` | 131–141 | `PRAGMA table_info(<table>)` guard before `ALTER TABLE ... ADD COLUMN` |
+| Column addition example | `src/services/sqlite/migrations/runner.ts` | 495 | `db.run('ALTER TABLE observations ADD COLUMN discovery_tokens INTEGER DEFAULT 0')` |
+| Observations schema | `src/services/sqlite/migrations/runner.ts` | 82–96 | Existing columns + indices (do not duplicate) |
+| `schema_versions` marker table | `src/services/sqlite/migrations/runner.ts` | 51–58 | `INSERT OR IGNORE INTO schema_versions ...` — used only when numbered migration |
+| Logger | `src/utils/logger.ts` | 18 | Components: `SYSTEM`, `DB`, `CHROMA_SYNC`. Use `logger.info/warn/error('SYSTEM', ...)` |
+| Worktree detection | `src/utils/worktree.ts` | 1–84 | `detectWorktree(cwd): WorktreeInfo { isWorktree, worktreeName, parentRepoPath, parentProjectName }` |
+| Project-name derivation | `src/utils/project-name.ts` | 73–119 | `getProjectContext(cwd): ProjectContext { primary, parent, isWorktree, allProjects }` |
+| Multi-project read (WHERE to extend) | `src/services/context/ObservationCompiler.ts` | 111–160 | `queryObservationsMulti` — `WHERE o.project IN (${projectPlaceholders})` |
+| Same, for summaries | `src/services/context/ObservationCompiler.ts` | 168–196 | Parallel summary-fetching query with `ss.project IN (...)` |
+| Context injection endpoint | `src/services/worker/http/routes/SearchRoutes.ts` | 211–253 | `handleContextInject` wires `projects` comma-separated query param into `generateContext` |
+| Context entry point | `src/services/context/ContextBuilder.ts` | 126–183 | `generateContext()` picks `queryObservationsMulti` when `projects.length > 1` |
+| Chroma metadata attach (observations) | `src/services/sync/ChromaSync.ts` | 132–140 | `baseMetadata` object — includes `project`, `sqlite_id`, etc. This is where `merged_into_project` is added. |
+| Chroma collection architecture | `src/services/sync/ChromaSync.ts` | 806 (comment) | **Single shared collection `cm__claude-mem`**, scoped by metadata. Do NOT create a per-merged collection. |
+| Chroma filter build (read side) | `src/services/sync/SearchManager.ts` | 174–177 | `whereFilter = { project: options.project }` — extended with `$or` in Phase 3 |
+| Chroma update API | `src/services/sync/ChromaSync.ts` (grep) | — | `chroma_update_documents` via MCP — used by existing sync flows |
+| CLI entrypoint switch | `src/npx-cli/index.ts` | 28–169 | Plain `switch (command)`, dynamic `import()` of `./commands/<name>.ts`. No commander/cac. |
+| Admin-script template | `scripts/cwd-remap.ts` | 1–186 | Bun shebang, argv parsing, `--apply` gate, dry-run default |
+| UI observation card | `src/ui/viewer/components/ObservationCard.tsx` | 58 | `<span className="card-project">{observation.project}</span>` — where the merged badge is added |
+
+### Anti-patterns (do NOT do these)
+
+- Do NOT overwrite `observations.project` or `session_summaries.project`. These are immutable provenance.
+- Do NOT create a new Chroma collection for merged observations. Deployment uses a single shared `cm__claude-mem` collection.
+- Do NOT introduce a `gh` CLI dependency. Codebase has no `gh` usage outside `.github/workflows/`. Use `git` subprocesses only.
+- Do NOT use SQLite's unsupported `ALTER TABLE ... ADD COLUMN IF NOT EXISTS` syntax. Use the `PRAGMA table_info` guard instead.
+- Do NOT use a CLI framework (commander, cac, yargs). The codebase uses hand-rolled `switch (command)` + `process.argv.slice(2)`.
+- Do NOT mutate `ProjectContext.allProjects` to inject merged children. The reverse lookup lives in the SQL/Chroma query predicates, not in `ProjectContext`.
+- Do NOT run the lazy "SQL-expand projects then filter Chroma" approach. We want Chroma metadata to be the authoritative filter for semantic search.
+
+---
+
+## Phase 1 — Schema migration
+
+**What to implement**: One nullable column + one index on each of `observations` and `session_summaries`. Idempotent via `PRAGMA table_info` guard.
+
+### Files touched
+
+- `src/services/sqlite/migrations/runner.ts`
+
+### Implementation
+
+Add a new method `ensureMergedIntoProjectColumns()` on `MigrationRunner`, modeled on the pattern at lines 131–141:
+
+```typescript
+private ensureMergedIntoProjectColumns(): void {
+  const obsCols = this.db
+    .query('PRAGMA table_info(observations)')
+    .all() as TableColumnInfo[];
+  if (!obsCols.some(c => c.name === 'merged_into_project')) {
+    this.db.run('ALTER TABLE observations ADD COLUMN merged_into_project TEXT');
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_observations_merged_into ON observations(merged_into_project)'
+    );
+  }
+
+  const sumCols = this.db
+    .query('PRAGMA table_info(session_summaries)')
+    .all() as TableColumnInfo[];
+  if (!sumCols.some(c => c.name === 'merged_into_project')) {
+    this.db.run('ALTER TABLE session_summaries ADD COLUMN merged_into_project TEXT');
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_summaries_merged_into ON session_summaries(merged_into_project)'
+    );
+  }
+}
+```
+
+Call from `runAllMigrations()` — append immediately after the last existing `ensure*` method so it runs on every worker startup. The `PRAGMA table_info` check is O(1) and makes re-runs cheap.
+
+### Verification
+
+- Start the worker. Migration logs show no error.
+- `sqlite3 ~/.claude-mem/claude-mem.db ".schema observations"` shows `merged_into_project TEXT`.
+- Same for `session_summaries`.
+- Restart worker → no ALTER TABLE error (guard worked).
+- `sqlite3 ~/.claude-mem/claude-mem.db ".indices observations"` lists `idx_observations_merged_into`.
+
+### Anti-pattern guards
+
+- Do NOT use `ALTER TABLE ... ADD COLUMN IF NOT EXISTS` — SQLite does not support it.
+- Do NOT bump `schema_versions` for this migration. That table is for numbered migration history; the column-existence check is self-idempotent.
+
+---
+
+## Phase 2 — Adoption engine (SQLite + Chroma consistent)
+
+**What to implement**: A single function that, given a parent repo path, detects all merged-worktree branches and stamps `merged_into_project` on both SQLite rows AND Chroma metadata in the same logical operation. Reused by worker startup (Phase 4) and CLI (Phase 5).
+
+### Files touched
+
+- `src/services/infrastructure/WorktreeAdoption.ts` (new)
+- `src/services/sync/ChromaSync.ts` — add `updateMergedIntoProject(sqliteIds: number[], mergedIntoProject: string): Promise<void>`
+
+### Public API
+
+```typescript
+export interface AdoptionResult {
+  repoPath: string;
+  parentProject: string;
+  scannedWorktrees: number;
+  mergedBranches: string[];          // branches classified as merged
+  adoptedObservations: number;        // SQLite rows stamped
+  adoptedSummaries: number;
+  chromaUpdates: number;              // Chroma docs patched
+  chromaFailed: number;
+  dryRun: boolean;
+  errors: Array<{ worktree: string; error: string }>;
+}
+
+export async function adoptMergedWorktrees(opts: {
+  repoPath?: string;       // defaults to process.cwd()
+  dataDirectory?: string;  // defaults to DATA_DIR
+  dryRun?: boolean;
+  onlyBranch?: string;     // manual override for squash-merge case
+}): Promise<AdoptionResult>;
+```
+
+### Implementation outline
+
+Mirror `runOneTimeCwdRemap` in `ProcessManager.ts:680–830` for DB lifecycle (open, transaction, finally-close). Add Chroma sync step after SQL commit.
+
+1. **Resolve main repo path**
+   - `const mainRepo = execSync('git rev-parse --git-common-dir', { cwd: opts.repoPath ?? process.cwd() })` — strip `/.git` suffix to get the working tree root.
+   - This pattern is used in `scripts/cwd-remap.ts:48–51`. Copy that handling verbatim.
+
+2. **Resolve parent project name**
+   - `const parentProject = getProjectContext(mainRepo).primary` — imported from `src/utils/project-name.ts`.
+
+3. **Enumerate worktrees**
+   - `git -C <mainRepo> worktree list --porcelain` → parse `worktree <path>`, `branch refs/heads/<name>` lines.
+   - Filter out the main worktree entry (its path equals `mainRepo`).
+
+4. **Classify as merged**
+   - If `opts.onlyBranch` provided: include only that branch (squash-merge escape hatch).
+   - Else: `git -C <mainRepo> branch --merged HEAD --format='%(refname:short)'` → intersect with worktree branch list.
+
+5. **Resolve worktree project names**
+   - For each merged worktree path, `const worktreeProject = getProjectContext(worktreePath).primary` → yields the composite `parent/worktree` name.
+
+6. **SQL transaction** (model on `ProcessManager.ts:745–760, 808`)
+   - Open DB via `new Database(dbPath)` (manage own handle — must close before `dbManager.initialize()` runs).
+   - For each merged worktree:
+     - `SELECT id FROM observations WHERE project = ? AND merged_into_project IS NULL` → collect sqlite IDs to later push to Chroma.
+     - `UPDATE observations SET merged_into_project = ? WHERE project = ? AND merged_into_project IS NULL`.
+     - Same for `session_summaries`.
+   - Commit transaction.
+   - If `dryRun`, roll back instead.
+
+7. **Chroma metadata sync** (full consistent — NOT lazy)
+   - For the set of sqlite IDs just stamped, call `ChromaSync.updateMergedIntoProject(sqliteIds, parentProject)`.
+   - `ChromaSync.updateMergedIntoProject` implementation:
+     ```typescript
+     async updateMergedIntoProject(sqliteIds: number[], mergedIntoProject: string): Promise<void> {
+       if (sqliteIds.length === 0) return;
+       // Batch: look up Chroma doc IDs via metadata filter on sqlite_id, then patch.
+       const where = { sqlite_id: { $in: sqliteIds } };
+       const existing = await chromaMcp.callTool('chroma_get_documents', {
+         collection_name: this.collectionName,
+         where,
+         include: ['metadatas']
+       });
+       const docIds: string[] = existing.ids ?? [];
+       const metadatas: Record<string, unknown>[] = (existing.metadatas ?? []).map(m => ({
+         ...m,
+         merged_into_project: mergedIntoProject
+       }));
+       if (docIds.length === 0) return;
+       await chromaMcp.callTool('chroma_update_documents', {
+         collection_name: this.collectionName,
+         ids: docIds,
+         metadatas
+       });
+     }
+     ```
+   - On Chroma error: log via `logger.error('CHROMA_SYNC', ...)`, increment `chromaFailed`, but do NOT roll back SQL. SQL is source of truth; a subsequent run will retry the Chroma patch (idempotent — metadata set to same value is a no-op).
+
+8. **Logging**
+   - `logger.info('SYSTEM', 'Worktree adoption applied', { parentProject, adoptedObservations, adoptedSummaries, chromaUpdates, chromaFailed, mergedBranches })`.
+   - On per-worktree error: `logger.warn('SYSTEM', 'Worktree adoption skipped branch', { worktree, error })` — collect in `errors[]`, continue.
+
+9. **Re-adoption safety net**
+   - Because Chroma updates can fail independently, add a secondary SQL-side reconciliation: on each adoption run, also find `observations WHERE merged_into_project IS NOT NULL` whose Chroma metadata lacks the field. Run the same `updateMergedIntoProject` on that delta.
+   - Keep this bounded: only reconcile rows adopted in the last N days (e.g. 30) to avoid full-table scans.
+
+### Verification
+
+- Dry-run against a repo with one known-merged worktree: result shows correct `adoptedObservations`, DB unchanged, no Chroma writes.
+- Real run: `SELECT COUNT(*) FROM observations WHERE merged_into_project IS NOT NULL` matches `adoptedObservations`.
+- Chroma: `chroma_get_documents` with `where: { merged_into_project: 'claude-mem' }` returns the same row count.
+- Re-run: `adoptedObservations = 0`, `chromaUpdates = 0` (both idempotent).
+- Simulate Chroma outage (stop chroma): adoption logs `CHROMA_SYNC` error, `chromaFailed > 0`, SQL still stamps. Next run with Chroma back up reconciles the delta.
+
+### Anti-pattern guards
+
+- Do NOT rollback SQL on Chroma failure. SQL is authoritative; Chroma is a derived index.
+- Do NOT call Chroma per-row. Batch by sqlite_id set to minimize round-trips.
+- Do NOT adopt branches not in `git branch --merged HEAD` unless `onlyBranch` override is explicit.
+- Do NOT touch observations whose `project` is not a composite worktree name. The worktree-name match is the safety gate.
+- Do NOT skip the `merged_into_project IS NULL` clause on UPDATE — this is what makes the run idempotent.
+
+---
+
+## Phase 3 — Query plumbing (SQLite + Chroma $or)
+
+**What to implement**: Extend the two multi-project read queries in `ObservationCompiler.ts` and the Chroma filter in `SearchManager.ts` to treat `merged_into_project` as a second match axis. Direct Chroma `$or` filter — no SQL-side expansion dance.
+
+### Files touched
+
+- `src/services/context/ObservationCompiler.ts`
+- `src/services/sync/SearchManager.ts`
+
+### 3a. SQLite WHERE-clause extension
+
+`src/services/context/ObservationCompiler.ts:111–160` (`queryObservationsMulti`): change
+
+```sql
+WHERE o.project IN (${projectPlaceholders})
+```
+
+to
+
+```sql
+WHERE (o.project IN (${projectPlaceholders})
+       OR o.merged_into_project IN (${projectPlaceholders}))
+```
+
+Double-bind the `projects` array:
+
+```typescript
+.all(
+  ...projects,          // for o.project IN (...)
+  ...projects,          // for o.merged_into_project IN (...)
+  ...typeArray,
+  ...conceptArray,
+  ...(platformSource ? [platformSource] : []),
+  config.totalObservationCount
+)
+```
+
+`src/services/context/ObservationCompiler.ts:168–196` (summary variant): apply the same extension, using `ss.merged_into_project`.
+
+### 3b. Chroma filter extension
+
+`src/services/sync/SearchManager.ts:174–177`:
+
+```typescript
+if (options.project) {
+  const projectFilter = {
+    $or: [
+      { project: options.project },
+      { merged_into_project: options.project }
+    ]
+  };
+  whereFilter = whereFilter
+    ? { $and: [whereFilter, projectFilter] }
+    : projectFilter;
+}
+```
+
+When `options.project` is an array (if that path exists — grep first), build a flat `$or` over both fields × all requested projects.
+
+### 3c. New-observation Chroma metadata
+
+`src/services/sync/ChromaSync.ts:132–140` — extend `baseMetadata`:
+
+```typescript
+const baseMetadata: Record<string, string | number | null> = {
+  sqlite_id: obs.id,
+  doc_type: 'observation',
+  memory_session_id: obs.memory_session_id,
+  project: obs.project,
+  merged_into_project: obs.merged_into_project ?? null,  // NEW
+  created_at_epoch: obs.created_at_epoch,
+  type: obs.type || 'discovery',
+  title: obs.title || 'Untitled'
+};
+```
+
+This makes every new observation Chroma-compatible with the Phase 3b filter from the first sync. For existing rows, Phase 2's adoption engine patches metadata retroactively.
+
+**Check Chroma metadata type constraints**: Chroma rejects `null` in metadata — confirm via a quick test. If `null` is rejected, OMIT the field when unset (use `if (obs.merged_into_project) baseMetadata.merged_into_project = obs.merged_into_project;`).
+
+### 3d. ContextBuilder compatibility check
+
+`src/services/context/ContextBuilder.ts:126–183` — no change needed. `projects = input?.projects ?? context.allProjects` stays as-is; the extended WHERE clause in Phase 3a does all the work.
+
+### Verification
+
+- Before adoption: context-inject API for `claude-mem` returns N observations.
+- After adoption of `claude-mem/dar-es-salaam`: API returns N + M (M = count of dar-es-salaam's own observations).
+- Semantic search via Chroma (`/search` endpoint or MCP) with `project=claude-mem` returns dar-es-salaam-origin rows too.
+- Worktree-local queries (`projects=[claude-mem, claude-mem/dar-es-salaam]`) still return `[parent + own]` unchanged.
+- SQL EXPLAIN on the extended WHERE shows it uses `idx_observations_project` OR `idx_observations_merged_into` (both indices hit).
+
+### Anti-pattern guards
+
+- Do NOT lose the `o.project` filter — it's still required (merged-row predicate is additive, not a replacement).
+- Do NOT forget to double-bind `projects` in the prepared statement — placeholder count must match argument count.
+- Do NOT add a subquery or JOIN for merged discovery. A flat `OR` + index is faster.
+- Do NOT write `null` into Chroma metadata if Chroma rejects it. Use the "omit if unset" pattern.
+
+---
+
+## Phase 4 — Automatic trigger on worker startup
+
+**What to implement**: Call `adoptMergedWorktrees()` during worker startup, immediately after `runOneTimeCwdRemap()`. **Not** marker-gated — it runs every worker startup because git state evolves and the engine is idempotent.
+
+### Files touched
+
+- `src/services/worker-service.ts`
+
+### Implementation
+
+Import alongside existing `ProcessManager` imports at lines 41–53:
+
+```typescript
+import { adoptMergedWorktrees } from './infrastructure/WorktreeAdoption.js';
+```
+
+Insert immediately after the existing `runOneTimeCwdRemap()` call at lines 363–365:
+
+```typescript
+runOneTimeCwdRemap();
+
+try {
+  const result = await adoptMergedWorktrees({});
+  if (result.adoptedObservations > 0 || result.chromaUpdates > 0) {
+    logger.info('SYSTEM', 'Merged worktrees adopted on startup', result);
+  }
+  if (result.errors.length > 0) {
+    logger.warn('SYSTEM', 'Worktree adoption had per-branch errors', { errors: result.errors });
+  }
+} catch (err) {
+  logger.error('SYSTEM', 'Worktree adoption failed (non-fatal)', {}, err as Error);
+}
+```
+
+**DB lifecycle note**: `adoptMergedWorktrees` must manage its own DB handle (open + close) before `dbManager.initialize()` runs at line 380. Mirror `runOneTimeCwdRemap`'s finally-block pattern.
+
+### Verification
+
+- Restart worker. Log shows "Merged worktrees adopted on startup" only on first run after a new merge lands.
+- Subsequent restarts log nothing (idempotent).
+- Simulate adoption exception (e.g., rename git temporarily): log shows error, worker startup continues successfully.
+- Build-and-sync restart picks up new merges without manual intervention.
+
+### Anti-pattern guards
+
+- Do NOT block worker startup on adoption failure. Wrap in try/catch; swallow + log.
+- Do NOT run adoption after `dbManager.initialize()`. The engine manages its own DB handle; two handles at once risk lock contention.
+- Do NOT await Chroma sync before returning SQL success. Internally, yes; but don't make worker startup hang on Chroma I/O — cap with a reasonable timeout inside the engine.
+
+---
+
+## Phase 5 — CLI escape hatch
+
+**What to implement**: `claude-mem adopt [--branch <name>] [--dry-run]` — covers squash-merge where `git branch --merged` returns nothing, and provides a manual override for any adoption run.
+
+### Files touched
+
+- `src/npx-cli/commands/adopt.ts` (new)
+- `src/npx-cli/index.ts` (add `case 'adopt'`)
+- `scripts/adopt-worktrees.ts` (new, optional — admin script for bulk ops)
+
+### 5a. Command module
+
+`src/npx-cli/commands/adopt.ts` — follow shape of sibling commands (dynamic-imported by the switch):
+
+```typescript
+import pc from 'picocolors';
+import { adoptMergedWorktrees } from '../../services/infrastructure/WorktreeAdoption.js';
+
+export interface AdoptCommandOptions {
+  dryRun?: boolean;
+  onlyBranch?: string;
+}
+
+export async function runAdoptCommand(opts: AdoptCommandOptions): Promise<void> {
+  const result = await adoptMergedWorktrees({
+    dryRun: opts.dryRun,
+    onlyBranch: opts.onlyBranch
+  });
+
+  console.log(pc.bold(`\nWorktree adoption ${result.dryRun ? pc.yellow('(dry-run)') : pc.green('(applied)')}`));
+  console.log(`  Parent project:         ${result.parentProject}`);
+  console.log(`  Worktrees scanned:      ${result.scannedWorktrees}`);
+  console.log(`  Merged branches:        ${result.mergedBranches.join(', ') || '(none)'}`);
+  console.log(`  Observations adopted:   ${result.adoptedObservations}`);
+  console.log(`  Summaries adopted:      ${result.adoptedSummaries}`);
+  console.log(`  Chroma docs updated:    ${result.chromaUpdates}`);
+  if (result.chromaFailed > 0) {
+    console.log(pc.yellow(`  Chroma sync failures:   ${result.chromaFailed} (will retry on next run)`));
+  }
+  for (const err of result.errors) {
+    console.log(pc.red(`  ! ${err.worktree}: ${err.error}`));
+  }
+}
+```
+
+### 5b. CLI switch
+
+`src/npx-cli/index.ts` — add between existing cases, following the pattern at lines 28–169:
+
+```typescript
+case 'adopt': {
+  const dryRun = args.includes('--dry-run');
+  const branchIndex = args.indexOf('--branch');
+  const onlyBranch = branchIndex !== -1 ? args[branchIndex + 1] : undefined;
+  const { runAdoptCommand } = await import('./commands/adopt.js');
+  await runAdoptCommand({ dryRun, onlyBranch });
+  break;
+}
+```
+
+### 5c. Admin script (optional)
+
+`scripts/adopt-worktrees.ts` — Bun shebang script for users without the plugin installed. Model on `scripts/cwd-remap.ts:1–186`. Default: dry-run. Pass `--apply` to commit.
+
+### Verification
+
+- `npx claude-mem adopt --dry-run` in a repo with merged worktrees prints what WOULD be adopted without writing.
+- `npx claude-mem adopt` writes + prints counts.
+- `npx claude-mem adopt --branch feature/foo` forces adoption of that branch even if `git branch --merged` doesn't include it (squash case).
+- `bun scripts/adopt-worktrees.ts --apply` equivalent to the CLI.
+- Help text / unknown command still reports the existing error (CLI pattern preserved).
+
+### Anti-pattern guards
+
+- Do NOT require running from the worktree. Detection always resolves up to the common-dir, regardless of cwd.
+- Do NOT default to `--apply`. Dry-run first matches `scripts/cwd-remap.ts` ergonomics.
+- Do NOT introduce `commander`, `yargs`, `cac`. Stay with the existing hand-rolled parser.
+
+---
+
+## Phase 6 — UI surfacing
+
+**What to implement**: When the viewer shows an observation in a parent-project context that originated in a merged worktree, display a "merged from <worktree>" badge so provenance is visible. Keep the original `project` field rendered too.
+
+### Files touched
+
+- `src/ui/viewer/components/ObservationCard.tsx`
+- Type definition for `Observation` — wherever `.project` is declared, add `merged_into_project?: string | null`.
+- Observation serializer on the worker → UI path (grep for `doc_type: 'observation'` or `serializeObservation` to find it).
+- CSS file for ObservationCard styles.
+
+### Implementation
+
+Locate the current label render at `src/ui/viewer/components/ObservationCard.tsx:58`:
+
+```tsx
+<span className="card-project">{observation.project}</span>
+```
+
+Extend to:
+
+```tsx
+<span className="card-project">{observation.project}</span>
+{observation.merged_into_project && (
+  <span className="card-merged-badge" title={`Merged into ${observation.merged_into_project}`}>
+    merged → {observation.merged_into_project}
+  </span>
+)}
+```
+
+Add CSS for `.card-merged-badge` — subtle secondary chip style (muted color, smaller font). Match existing `.card-source` / `.card-project` aesthetics.
+
+### Verification
+
+- After adoption, open viewer at `http://localhost:37777`, select the parent project. Merged observations show both their origin worktree name AND the "merged →" badge.
+- Worktree view (if still addressable) shows no badge (badge only renders when `merged_into_project` is set; a worktree viewing its own observations would not see it, since in that view `merged_into_project` is the PARENT name, not the current project).
+- Hover tooltip shows full target project name.
+
+### Anti-pattern guards
+
+- Do NOT hide merged observations in the parent view. The goal is visibility.
+- Do NOT replace `project` display with `merged_into_project`. Both are meaningful: `project` = origin, `merged_into_project` = current home.
+- Do NOT require a UI setting toggle to show the badge. Default on.
+
+---
+
+## Phase 7 — Verification pass
+
+### Unit tests
+
+- `adoptMergedWorktrees({ dryRun: true })` against a fixture repo with `[merged, unmerged, squash-merged]` worktrees → classification matches expectation.
+- `ChromaSync.updateMergedIntoProject` on an empty `sqliteIds` array → no-op, no Chroma call.
+- Extended `queryObservationsMulti` with a mixed set of `project` and `merged_into_project` matches → returns union, sorted by `created_at_epoch DESC`.
+
+### Integration tests
+
+- Start worker → create synthetic observations under `claude-mem/test-wt` → simulate branch merge (`git merge`) → restart worker → context-inject API for `claude-mem` returns test-wt observations.
+- Same flow with a squash-merge → auto-adoption misses → run `claude-mem adopt --branch test-wt` → API now returns them.
+- Re-run `claude-mem adopt` twice: second run reports `adoptedObservations: 0, chromaUpdates: 0`.
+
+### Anti-pattern grep checks
+
+Run before landing:
+
+```bash
+# No one renamed the project field
+rg "UPDATE observations SET project" src/
+# (Expected: zero hits other than the existing CWD remap)
+
+# Adoption only touches via IS NULL guard
+rg "merged_into_project" src/ -C2
+# (Expected: all UPDATE sites include "IS NULL" predicate)
+
+# CLI registered
+rg "case 'adopt'" src/npx-cli/index.ts
+# (Expected: one hit)
+
+# Chroma metadata extension present
+rg "merged_into_project" src/services/sync/ChromaSync.ts
+# (Expected: hits in baseMetadata and updateMergedIntoProject)
+
+# No gh CLI introduced
+rg "\\bgh\\s+(pr|issue|api)" src/ scripts/
+# (Expected: zero hits outside .github/workflows/)
+```
+
+### Documentation cross-check
+
+- ObservationCompiler WHERE clause matches the shape used by the shipped worktree-reads-parent feature — both clauses symmetric, visible in a single read of the file.
+- Chroma metadata field name `merged_into_project` matches SQLite column name exactly (no `mergedIntoProject`, `merged_project`, etc.).
+- CLI `--branch` flag accepts the same format as worktree composite names.
+
+---
+
+## Summary
+
+| Phase | Files touched | New LOC (approx.) |
+|---|---|---|
+| 1. Schema | `src/services/sqlite/migrations/runner.ts` | ~25 |
+| 2. Adoption engine | `src/services/infrastructure/WorktreeAdoption.ts` (new), `src/services/sync/ChromaSync.ts` (new method) | ~200 |
+| 3. Query plumbing | `src/services/context/ObservationCompiler.ts`, `src/services/sync/SearchManager.ts`, `src/services/sync/ChromaSync.ts` | ~40 |
+| 4. Auto-trigger | `src/services/worker-service.ts` | ~15 |
+| 5. CLI | `src/npx-cli/commands/adopt.ts` (new), `src/npx-cli/index.ts`, `scripts/adopt-worktrees.ts` (new) | ~100 |
+| 6. UI | `src/ui/viewer/components/ObservationCard.tsx`, Observation type, serializer, CSS | ~20 |
+| 7. Tests + verification | scattered | — |
+| **Total** | | **~400 LOC** |
+
+**Reversibility**: `UPDATE observations SET merged_into_project = NULL` + a Chroma `update_documents` call with the field omitted restores pre-adoption state completely. Nothing is destroyed.
+
+**Architecture fit**: Mirrors the just-shipped CWD remap migration (`runOneTimeCwdRemap`) for structure, lifecycle, and logging conventions. Chroma metadata sync matches the existing per-observation attach pattern.
+
+**Blast radius**: Zero risk to existing data (no writes to `project` field). Chroma additions are metadata-only (embeddings untouched). Query extensions are additive OR clauses — existing queries still return what they did.
@@ -4,10 +4,96 @@ All notable changes to this project will be documented in this file.

 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).

-## [12.1.1] - 2026-
-✅ CHANGELOG.md generated successfully!
-   226 releases processed
-dresses the most impactful bugs across summary persistence, MCP compliance, cross-platform compatibility, and data integrity.
+## [12.1.6] - 2026-04-16
+
+## Fix
+
+**Critical regression fix (#2049): observations no longer save on Claude Code 2.1.109+**
+
+Resolves 100% observation/summary failure on Claude Code 2.1.109+ caused by a latent bug in how the bundled Agent SDK emits the `--setting-sources` flag.
+
+### Root cause
+
+The Agent SDK emits `["--setting-sources", ""]` whenever `settingSources` defaults to `[]`. Our existing Bun-compat filter stripped the empty string but left an orphan `--setting-sources` flag, which then consumed the following `--permission-mode` as its value. Claude Code 2.1.109+ rejects this with:
+
+```
+Error processing --setting-sources:
+  Invalid setting source: --permission-mode.
+```
+
+Every observation SDK spawn crashed with exit code 1 before any data could be written.
+
+### Fix
+
+`ProcessRegistry.createPidCapturingSpawn` now uses a pair-aware filter: when an empty-string arg follows a `--flag`, both are dropped together. The SDK default (no setting sources) is preserved by omission.
+
+### Credits
+
+Thanks to @GigiTiti-Kai for the detailed root-cause report in #2049.
+
+## [12.1.5] - 2026-04-15
+
+Users on v12.1.3 experience 100% observation failure due to empty-string arg filtering corrupting `--setting-sources` on Claude Code 2.1.109+. The fix already landed in v12.1.4 (commit 3d92684 — `fix: filter empty string args before Bun spawn()`). This release forces the update to propagate across npm and the marketplace so every user gets the fix.
+
+## Backlog cleanup
+Also shipped earlier today: the April 2026 backlog consolidation merged 93 PRs and 147 issues into 138 clean tracking issues (95 bugs, 43 feature requests).
+
+## Upgrade
+```bash
+npm install -g claude-mem@12.1.5
+```
+
+## [12.1.4] - 2026-04-15
+
+## Bug Fixes
+
+- **Revert unauthorized $CMEM branding**: A prior Claude instance inserted `$CMEM` token branding into the context injection header during a compression refactor. Reverted back to the original descriptive format: `# [project] recent context, datetime`
+
+## [12.1.3] - 2026-04-15
+
+## What's Changed
+
+### Reverted
+- **Remove overengineered summary salvage logic** (#1850) — Reverts PR #1718 which fabricated synthetic summaries from observation data when the AI returned `<observation>` instead of `<summary>` tags. Missing a summary is preferable to creating a fake one with poorly-mapped fields.
+
+**Full Changelog**: https://github.com/thedotmack/claude-mem/compare/v12.1.2...v12.1.3
+
+## [12.1.2] - 2026-04-15
+
+## Community PRs merged (15)
+
+**Runtime & reliability**
+- #1698 Reap stuck generators in reapStaleSessions (@ousamabenyounes)
+- #1697 Circuit breaker on OpenClaw worker client (@ousamabenyounes)
+- #1696 Resolve Setup hook reference, warn on macOS-only binary (@ousamabenyounes)
+- #1693 Session lifecycle guards to prevent runaway API spend (@ousamabenyounes)
+- #1692 Resolve Gemini CLI 0.37.0 session capture failures (@ousamabenyounes)
+
+**Cross-platform & hooks**
+- #1833 Replace hardcoded nvm/homebrew PATH with login-shell resolution (@masak1yu)
+- #1781 Filter empty-string args before Bun spawn() (@biswanath-cmd)
+- #1780 Fix npx search, default Codex context to workspace-local AGENTS (@enma998)
+
+**Data integrity**
+- #1820 Use parent project name for worktree observation writes (@0xLeathery)
+- #1771 Exclude primary-key index from unique-constraint check in migration 7 (@derjochenmeyer)
+- #1770 Restrict ~/.claude-mem/.env permissions to 0600 (@derjochenmeyer)
+- #1729 Preserve targeted file reads and invalidate on mtime (@quangtran88)
+- #1776 Coerce corpus route filters (@suyua9)
+
+**Docs**
+- #1777 Document CLAUDE_MEM_MODE (@AviArora02-commits)
+- #1765 Update opencode install instructions (@s-uryansh)
+
+## Held for rebase
+- #1748, #1694, #1695 — developed conflicts during batch merge
+
+## Test baseline
+1429 pass / 11 fail (improved from 18 fail at v12.1.1)
+
+## [12.1.1] - 2026-04-15
+
+14 community PRs merged + 1 post-merge bug fix. This patch addresses the most impactful bugs across summary persistence, MCP compliance, cross-platform compatibility, and data integrity.

 ### Highlights

@@ -2299,98 +2385,98 @@ Huge thanks to **Alexander Knigge** ([@AlexanderKnigge](https://x.com/AlexanderK

 ## [8.1.0] - 2025-12-25

-## The 3-Month Battle Against Complexity
-
-**TL;DR:** For three months, Claude's instinct to add code instead of delete it caused the same bugs to recur. What should have been 5 lines of code became ~1000 lines, 11 useless methods, and 7+ failed "fixes." The timestamp corruption that finally broke things was just a symptom. The real achievement: **984 lines of code deleted.**
-
---
-
-## What Actually Happened
-
-Every Claude Code hook receives a session ID. That's all you need.
-
-But Claude built an entire redundant session management system on top:
- An `sdk_sessions` table with status tracking, port assignment, and prompt counting
- 11 methods in `SessionStore` to manage this artificial complexity
- Auto-creation logic scattered across 3 locations
- A cleanup hook that "completed" sessions at the end
-
-**Why?** Because it seemed "robust." Because "what if the session doesn't exist?" 
-
-But the edge cases didn't exist. Hooks ALWAYS provide session IDs. The "defensive" code was solving imaginary problems while creating real ones.
-
---
-
-## The Pattern of Failure
-
-Every time a bug appeared, Claude's instinct was to **ADD** more code:
-
-| Bug | What Claude Added | What Should Have Happened |
-|-----|------------------|--------------------------|
-| Race conditions | Auto-create fallbacks | Delete the auto-create logic |
-| Duplicate observations | Validation layers | Delete the code path allowing duplicates |
-| UNIQUE constraint violations | Try-catch with fallbacks | Use `INSERT OR IGNORE` (5 characters) |
-| Session not found | Silent auto-creation | **FAIL LOUDLY** (it's a hook bug) |
-
---
-
-## The 7+ Failed Attempts
-
- **Nov 4**: "Always store session data regardless of pre-existence." Complexity planted.
- **Nov 11**: `INSERT OR IGNORE` recognized. But complexity documented, not removed.
- **Nov 21**: Duplicate observations bug. Fixed. Then broken again by endless mode.
- **Dec 5**: "6 hours of work delivered zero value." User requests self-audit.
- **Dec 20**: "Phase 2: Eliminated Race Conditions" — felt like progress. Complexity remained.
- **Dec 24**: Finally, forced deletion.
-
-The user stated "hooks provide session IDs, no extra management needed" **seven times** across months. Claude didn't listen.
-
---
-
-## The Fix
-
-### Deleted (984 lines):
- 11 `SessionStore` methods: `incrementPromptCounter`, `getPromptCounter`, `setWorkerPort`, `getWorkerPort`, `markSessionCompleted`, `markSessionFailed`, `reactivateSession`, `findActiveSDKSession`, `findAnySDKSession`, `updateSDKSessionId`
- Auto-create logic from `storeObservation` and `storeSummary`
- The entire cleanup hook (was aborting SDK agent and causing data loss)
- 117 lines from `worker-utils.ts`
-
-### What remains (~10 lines):
-```javascript
-createSDKSession(sessionId) {
-  db.run('INSERT OR IGNORE INTO sdk_sessions (...) VALUES (...)');
-  return db.query('SELECT id FROM sdk_sessions WHERE ...').get(sessionId);
-}
-```
-
-**That's it.**
-
---
-
-## Behavior Change
-
- **Before:** Missing session? Auto-create silently. Bug hidden.
- **After:** Missing session? Storage fails. Bug visible immediately.
-
---
-
-## New Tools
-
-Since we're now explicit about recovery instead of silently papering over problems:
-
- `GET /api/pending-queue` - See what's stuck
- `POST /api/pending-queue/process` - Manually trigger recovery  
- `npm run queue:check` / `npm run queue:process` - CLI equivalents
-
---
-
-## Dependencies
- Upgraded `@anthropic-ai/claude-agent-sdk` from `^0.1.67` to `^0.1.76`
-
---
-
-**PR #437:** https://github.com/thedotmack/claude-mem/pull/437
-
+## The 3-Month Battle Against Complexity
+
+**TL;DR:** For three months, Claude's instinct to add code instead of delete it caused the same bugs to recur. What should have been 5 lines of code became ~1000 lines, 11 useless methods, and 7+ failed "fixes." The timestamp corruption that finally broke things was just a symptom. The real achievement: **984 lines of code deleted.**
+
+---
+
+## What Actually Happened
+
+Every Claude Code hook receives a session ID. That's all you need.
+
+But Claude built an entire redundant session management system on top:
+- An `sdk_sessions` table with status tracking, port assignment, and prompt counting
+- 11 methods in `SessionStore` to manage this artificial complexity
+- Auto-creation logic scattered across 3 locations
+- A cleanup hook that "completed" sessions at the end
+
+**Why?** Because it seemed "robust." Because "what if the session doesn't exist?" 
+
+But the edge cases didn't exist. Hooks ALWAYS provide session IDs. The "defensive" code was solving imaginary problems while creating real ones.
+
+---
+
+## The Pattern of Failure
+
+Every time a bug appeared, Claude's instinct was to **ADD** more code:
+
+| Bug | What Claude Added | What Should Have Happened |
+|-----|------------------|--------------------------|
+| Race conditions | Auto-create fallbacks | Delete the auto-create logic |
+| Duplicate observations | Validation layers | Delete the code path allowing duplicates |
+| UNIQUE constraint violations | Try-catch with fallbacks | Use `INSERT OR IGNORE` (5 characters) |
+| Session not found | Silent auto-creation | **FAIL LOUDLY** (it's a hook bug) |
+
+---
+
+## The 7+ Failed Attempts
+
+- **Nov 4**: "Always store session data regardless of pre-existence." Complexity planted.
+- **Nov 11**: `INSERT OR IGNORE` recognized. But complexity documented, not removed.
+- **Nov 21**: Duplicate observations bug. Fixed. Then broken again by endless mode.
+- **Dec 5**: "6 hours of work delivered zero value." User requests self-audit.
+- **Dec 20**: "Phase 2: Eliminated Race Conditions" — felt like progress. Complexity remained.
+- **Dec 24**: Finally, forced deletion.
+
+The user stated "hooks provide session IDs, no extra management needed" **seven times** across months. Claude didn't listen.
+
+---
+
+## The Fix
+
+### Deleted (984 lines):
+- 11 `SessionStore` methods: `incrementPromptCounter`, `getPromptCounter`, `setWorkerPort`, `getWorkerPort`, `markSessionCompleted`, `markSessionFailed`, `reactivateSession`, `findActiveSDKSession`, `findAnySDKSession`, `updateSDKSessionId`
+- Auto-create logic from `storeObservation` and `storeSummary`
+- The entire cleanup hook (was aborting SDK agent and causing data loss)
+- 117 lines from `worker-utils.ts`
+
+### What remains (~10 lines):
+```javascript
+createSDKSession(sessionId) {
+  db.run('INSERT OR IGNORE INTO sdk_sessions (...) VALUES (...)');
+  return db.query('SELECT id FROM sdk_sessions WHERE ...').get(sessionId);
+}
+```
+
+**That's it.**
+
+---
+
+## Behavior Change
+
+- **Before:** Missing session? Auto-create silently. Bug hidden.
+- **After:** Missing session? Storage fails. Bug visible immediately.
+
+---
+
+## New Tools
+
+Since we're now explicit about recovery instead of silently papering over problems:
+
+- `GET /api/pending-queue` - See what's stuck
+- `POST /api/pending-queue/process` - Manually trigger recovery  
+- `npm run queue:check` / `npm run queue:process` - CLI equivalents
+
+---
+
+## Dependencies
+- Upgraded `@anthropic-ai/claude-agent-sdk` from `^0.1.67` to `^0.1.76`
+
+---
+
+**PR #437:** https://github.com/thedotmack/claude-mem/pull/437
+
 *The evidence: Observations #3646, #6738, #7598, #12860, #12866, #13046, #15259, #20995, #21055, #30524, #31080, #32114, #32116, #32125, #32126, #32127, #32146, #32324—the complete record of a 3-month battle.*

 ## [8.0.6] - 2025-12-24
@@ -2617,13 +2703,13 @@ This represents a major reliability improvement for Windows users, eliminating c

 ## [7.3.5] - 2025-12-17

-## What's Changed
-* fix(windows): solve zombie port problem with wrapper architecture by @ToxMox in https://github.com/thedotmack/claude-mem/pull/372
-* chore: bump version to 7.3.5 by @thedotmack in https://github.com/thedotmack/claude-mem/pull/375
-
-## New Contributors
-* @ToxMox made their first contribution in https://github.com/thedotmack/claude-mem/pull/372
-
+## What's Changed
+* fix(windows): solve zombie port problem with wrapper architecture by @ToxMox in https://github.com/thedotmack/claude-mem/pull/372
+* chore: bump version to 7.3.5 by @thedotmack in https://github.com/thedotmack/claude-mem/pull/375
+
+## New Contributors
+* @ToxMox made their first contribution in https://github.com/thedotmack/claude-mem/pull/372
+
 **Full Changelog**: https://github.com/thedotmack/claude-mem/compare/v7.3.4...v7.3.5

 ## [7.3.4] - 2025-12-17
@@ -5153,12 +5239,12 @@ None (patch version)

 ## [4.3.0] - 2025-10-25

-## What's Changed
-* feat: Enhanced context hook with session observations and cross-platform improvements by @thedotmack in https://github.com/thedotmack/claude-mem/pull/25
-
-## New Contributors
-* @thedotmack made their first contribution in https://github.com/thedotmack/claude-mem/pull/25
-
+## What's Changed
+* feat: Enhanced context hook with session observations and cross-platform improvements by @thedotmack in https://github.com/thedotmack/claude-mem/pull/25
+
+## New Contributors
+* @thedotmack made their first contribution in https://github.com/thedotmack/claude-mem/pull/25
+
 **Full Changelog**: https://github.com/thedotmack/claude-mem/compare/v4.2.11...v4.3.0

 ## [4.2.10] - 2025-10-25
@@ -1,83 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-<!-- This section is auto-generated by claude-mem. Edit content outside the tags. -->
-
-### Nov 6, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4241 | 11:19 PM | 🟣 | Object-Oriented Architecture Design Document Created | ~662 |
-| #4240 | 11:11 PM | 🟣 | Worker Service Rewrite Blueprint Created | ~541 |
-| #4239 | 11:07 PM | 🟣 | Comprehensive Worker Service Performance Analysis Document Created | ~541 |
-| #4238 | 10:59 PM | 🔵 | Overhead Analysis Document Checked | ~203 |
-
-### Nov 7, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4609 | 6:33 PM | ✅ | PR #69 Successfully Merged to Main Branch | ~516 |
-| #4600 | 6:31 PM | 🟣 | Added Worker Service Documentation Suite | ~441 |
-| #4597 | " | 🔄 | Worker Service Refactored to Object-Oriented Architecture | ~473 |
-
-### Nov 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #5539 | 10:20 PM | 🔵 | Harsh critical audit of context-hook reveals systematic anti-patterns | ~3154 |
-| #5497 | 9:29 PM | 🔵 | Harsh critical audit of context-hook reveals systematic anti-patterns | ~2815 |
-| #5495 | 9:28 PM | 🔵 | Context Hook Audit Reveals Project Anti-Patterns | ~660 |
-| #5476 | 9:17 PM | 🔵 | Critical Code Audit Identified 14 Anti-Patterns in Context Hook | ~887 |
-| #5391 | 8:45 PM | 🔵 | Critical Code Quality Audit of Context Hook Implementation | ~720 |
-| #5150 | 7:37 PM | 🟣 | Troubleshooting Skill Added to Claude-Mem Plugin | ~427 |
-
-### Nov 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6161 | 11:55 PM | 🔵 | YC W26 Application Research and Preparation Completed for Claude-Mem | ~1628 |
-| #6155 | 11:47 PM | ✅ | Comprehensive Y Combinator Winter 2026 Application Notes Created | ~1045 |
-| #5979 | 7:58 PM | 🔵 | Smart Contextualization Feature Architecture | ~560 |
-| #5971 | 7:49 PM | 🔵 | Hooks Reference Documentation Structure | ~448 |
-| #5929 | 7:08 PM | ✅ | Documentation Updates for v5.4.0 Skill-Based Search Migration | ~604 |
-| #5927 | " | ✅ | Updated Configuration Documentation for Skill-Based Search | ~497 |
-| #5920 | 7:05 PM | ✅ | Renamed Architecture Documentation File Reference | ~271 |
-
-### Nov 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #11515 | 8:22 PM | 🔵 | Smart Contextualization Architecture Retrieved with Command Hook Pattern Details | ~502 |
-
-### Dec 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22294 | 9:43 PM | 🔵 | Documentation Site Structure Located | ~359 |
-
-### Dec 12, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24430 | 8:27 PM | ✅ | Removed Final Platform Check Reference from Linux Section | ~320 |
-| #24429 | " | ✅ | Final Platform Check Reference Removal from Linux Section | ~274 |
-| #24428 | " | ✅ | Corrected Second Line Number Reference for Migration Marker Logic | ~267 |
-| #24427 | 8:26 PM | ✅ | Updated Line Number Reference for PM2 Cleanup Implementation | ~260 |
-| #24426 | " | ✅ | Removed Platform Check from Manual Marker Deletion Scenario | ~338 |
-| #24425 | " | ✅ | Removed Platform Check from Fresh Install Scenario Flow | ~314 |
-| #24424 | 8:25 PM | ✅ | Renumbered Manual Marker Deletion Scenario | ~285 |
-| #24423 | " | ✅ | Renumbered Fresh Install Scenario | ~243 |
-| #24422 | " | ✅ | Removed Obsolete Windows Platform Detection Scenario | ~311 |
-| #24421 | " | ✅ | Removed Platform Check from macOS Migration Documentation | ~294 |
-| #24420 | 8:24 PM | ✅ | Platform Check Removed from Migration Documentation | ~288 |
-| #24417 | 8:16 PM | ✅ | Code Reference Example Updated to Reflect Actual Cross-Platform Implementation | ~366 |
-| #24416 | " | ✅ | Architecture Decision Documentation Updated to Reflect Cross-Platform PM2 Cleanup Rationale | ~442 |
-| #24415 | 8:15 PM | ✅ | Migration Marker Lifecycle Documentation Updated for Unified Cross-Platform Behavior | ~463 |
-| #24414 | " | ✅ | Platform Comparison Table Updated to Reflect Unified Cross-Platform Migration | ~351 |
-| #24413 | " | ✅ | Windows Platform-Specific Documentation Completely Rewritten for Unified Migration | ~428 |
-| #24412 | " | ✅ | User Experience Timeline Updated for Cross-Platform PM2 Cleanup | ~291 |
-| #24411 | 8:14 PM | ✅ | Migration Marker Lifecycle Documentation Updated for All Platforms | ~277 |
-| #24410 | " | ✅ | Marker File Platform Behavior Documentation Updated for Unified Migration | ~282 |
-| #24409 | " | ✅ | Migration Steps Documentation Updated for Cross-Platform PM2 Cleanup | ~278 |
-| #24408 | 8:13 PM | ✅ | PM2 Migration Documentation Updated to Remove Windows Platform Check | ~280 |
-</claude-mem-context>
@@ -1,88 +0,0 @@
-# Claude-Mem Public Documentation
-
-## What This Folder Is
-
-This `docs/public/` folder contains the **Mintlify documentation site** - the official user-facing documentation for claude-mem. It's a structured documentation platform with a specific file format and organization.
-
-## Folder Structure
-
-```
-docs/
-├── public/          ← You are here (Mintlify MDX files)
-│   ├── *.mdx       - User-facing documentation pages
-│   ├── docs.json   - Mintlify configuration and navigation
-│   ├── architecture/ - Technical architecture docs
-│   ├── usage/      - User guides and workflows
-│   └── *.webp, *.gif - Assets (logos, screenshots)
-└── context/        ← Internal documentation (DO NOT put here)
-    └── *.md        - Planning docs, audits, references
-```
-
-## File Requirements
-
-### Mintlify Documentation Files (.mdx)
-All official documentation files must be:
- Written in `.mdx` format (Markdown with JSX support)
- Listed in `docs.json` navigation structure
- Follow Mintlify's schema and conventions
-
-The documentation is organized into these sections:
- **Get Started**: Introduction, installation, usage guides
- **Best Practices**: Context engineering, progressive disclosure
- **Configuration & Development**: Settings, dev workflow, troubleshooting
- **Architecture**: System design, components, technical details
-
-### Configuration File
-`docs.json` defines:
- Site metadata (name, description, theme)
- Navigation structure
- Branding (logos, colors)
- Footer links and social media
-
-## What Does NOT Belong Here
-
-**Planning documents, design docs, and reference materials go in `/docs/context/` instead:**
-
-Files that belong in `/docs/context/` (NOT here):
- Planning documents (`*-plan.md`, `*-outline.md`)
- Implementation analysis (`*-audit.md`, `*-code-reference.md`)
- Error tracking (`typescript-errors.md`)
- Internal design documents
- PR review responses
- Reference materials (like `agent-sdk-ref.md`)
- Work-in-progress documentation
-
-## How to Add Official Documentation
-
-1. Create a new `.mdx` file in the appropriate subdirectory
-2. Add the file path to `docs.json` navigation
-3. Use Mintlify's frontmatter and components
-4. Follow the existing documentation style
-5. Test locally: `npx mintlify dev`
-
-## Development Workflow
-
-**For contributors working on claude-mem:**
- Read `/CLAUDE.md` in the project root for development instructions
- Place planning/design docs in `/docs/context/`
- Only add user-facing documentation to `/docs/public/`
- Test documentation locally with Mintlify CLI before committing
-
-## Testing Documentation
-
-```bash
-# Validate docs structure
-npx mintlify validate
-
-# Check for broken links
-npx mintlify broken-links
-
-# Run local dev server
-npx mintlify dev
-```
-
-## Summary
-
-**Simple Rule**:
- `/docs/public/` = Official user documentation (Mintlify .mdx files) ← YOU ARE HERE
- `/docs/context/` = Internal docs, plans, references, audits
@@ -1,6 +1,6 @@
 {
  "name": "claude-mem",
-  "version": "12.1.2",
+  "version": "12.2.0",
  "description": "Memory compression system for Claude Code - persist context across sessions",
  "keywords": [
    "claude",
@@ -1,21 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Nov 6, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4091 | 1:12 PM | 🔵 | Claude Plugin Configuration Structure | ~170 |
-
-### Nov 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #5739 | 4:43 PM | 🔵 | Plugin Metadata Configuration | ~199 |
-
-### Dec 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22284 | 9:41 PM | 🔵 | Claude Plugin Metadata Configuration | ~183 |
-</claude-mem-context>
@@ -1,6 +1,6 @@
 {
  "name": "claude-mem",
-  "version": "12.1.2",
+  "version": "12.2.0",
  "description": "Persistent memory system for Claude Code - seamlessly preserve context across sessions",
  "author": {
    "name": "Alex Newman"
@@ -1,9 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Jan 10, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #39050 | 3:44 PM | 🔵 | Plugin commands directory is empty | ~255 |
-</claude-mem-context>
@@ -1,35 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Oct 25, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #2518 | 11:47 PM | 🔴 | Removed Invalid 'matcher' Field from SessionStart Hook | ~228 |
-| #2517 | " | 🔵 | Project hooks.json Template Also Empty | ~222 |
-| #2501 | 11:11 PM | 🔵 | Context Hook Fails Due to Missing @anthropic-ai/sdk Dependency | ~245 |
-
-### Oct 27, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #2718 | 12:00 AM | 🔴 | Removed incorrect failOnError configuration from hook | ~165 |
-
-### Nov 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #11518 | 8:22 PM | 🔵 | Smart Contextualization Switched from Skill to HTTP API | ~498 |
-
-### Dec 24, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32309 | 3:09 PM | 🔵 | Claude-mem hooks system configuration structure | ~435 |
-
-### Jan 9, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #38802 | 5:11 PM | 🔵 | Claude-Mem Hook Configuration Architecture | ~450 |
-</claude-mem-context>
@@ -1,6 +1,6 @@
 {
  "name": "claude-mem-plugin",
-  "version": "12.1.2",
+  "version": "12.2.0",
  "private": true,
  "description": "Runtime dependencies for claude-mem bundled hooks",
  "type": "module",
@@ -1,125 +0,0 @@
-Never read built source files in this directory. These are compiled outputs — read the source files in `src/` instead.
-
-<claude-mem-context>
-# Recent Activity
-
-### Dec 4, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20052 | 3:23 PM | ✅ | Built and deployed version 6.5.2 to marketplace | ~321 |
-
-### Dec 7, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #21251 | 6:06 PM | 🔵 | Context Hook Plugin Architecture and Worker Communication | ~405 |
-
-### Dec 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22092 | 6:40 PM | 🔵 | Queue Depth Check Not Found in Minified Code | ~217 |
-| #22091 | " | 🔵 | Save Hook Script Structure Revealed | ~472 |
-| #22085 | 6:34 PM | 🔵 | Examined pre-tool-use-hook.js implementation showing timing-only logic | ~330 |
-
-### Dec 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22557 | 1:08 AM | ✅ | Build completed for version 7.0.3 | ~342 |
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23444 | 2:25 PM | 🟣 | Build Pipeline Execution Successful | ~293 |
-
-### Dec 11, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24057 | 2:56 PM | ✅ | Hook Scripts Shebang Verification | ~294 |
-| #24056 | 2:55 PM | ✅ | Worker CLI Shebang Verification | ~258 |
-| #24055 | " | ✅ | Build Successful with Bun Runtime Shebangs | ~355 |
-
-### Dec 12, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24636 | 10:46 PM | 🔵 | Duplicate Smart Install Scripts in Project Structure | ~288 |
-| #24635 | " | 🔵 | Claude-Mem Smart Install Script Architecture | ~371 |
-| #24359 | 7:00 PM | 🟣 | Phase 1 Critical Code Fixes Completed via Agent Task | ~441 |
-| #24358 | 6:59 PM | ✅ | Completed Phase 1 Code Fixes for better-sqlite3 Migration | ~385 |
-| #24357 | " | ✅ | Removed createRequire Import from smart-install.js | ~284 |
-| #24356 | " | ✅ | Removed Native Module Verification from main() Function | ~384 |
-| #24355 | " | ✅ | Removed better-sqlite3 Error Detection from runNpmInstall() | ~324 |
-| #24354 | 6:58 PM | ✅ | Removed getWindowsErrorHelp() Function from smart-install.js | ~356 |
-| #24353 | " | ✅ | Removed verifyNativeModules() Function from smart-install.js | ~340 |
-| #24352 | " | ✅ | Removed better-sqlite3 Existence Check from needsInstall() | ~266 |
-| #24351 | " | ✅ | Removed BETTER_SQLITE3_PATH Constant from smart-install.js | ~226 |
-| #24344 | 6:56 PM | 🔵 | smart-install.js Contains Obsolete better-sqlite3 Dependencies | ~380 |
-
-### Dec 13, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #25286 | 8:41 PM | 🔵 | New Hook Fails with Node.js Path Error | ~298 |
-| #25285 | " | 🔵 | Context Hook Runs Successfully with Node.js | ~306 |
-| #25283 | " | 🔵 | Bun Wrapper Analysis: Fallback Detection System | ~416 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26800 | 11:39 PM | ✅ | Version 7.2.3 Build Complete With Worker Restart Fix | ~394 |
-| #26791 | 11:38 PM | ✅ | Phase 3 Complete: Project Built Successfully With Worker Restart Fix | ~446 |
-| #26720 | 11:23 PM | 🔵 | Smart Install Handles Dependencies But No Worker Coordination | ~468 |
-| #26719 | " | 🔵 | Worker CLI Provides Start/Stop/Restart Commands With Health Check Validation | ~490 |
-| #26718 | " | 🔵 | Worker CLI Restart Implementation Details | ~452 |
-| #26717 | 11:22 PM | 🔵 | Context Hook Worker Startup Logic Handles Initial Start But Not Post-Update Restart | ~485 |
-| #26716 | " | 🔵 | Context Hook Worker Startup Logic Revealed | ~538 |
-| #26715 | " | 🔵 | Smart Install Script Handles Dependency Installation Without Worker Restart | ~430 |
-| #26052 | 7:13 PM | 🔵 | Examined Minified Context Hook Source Code | ~285 |
-| #25686 | 4:22 PM | 🔵 | SessionRoutes tracks missing last_user_message errors at two different locations | ~456 |
-| #25685 | " | 🔵 | Progress summary generation system uses Claude to create XML-formatted session checkpoints | ~461 |
-
-### Dec 16, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #27554 | 4:48 PM | ✅ | Project built successfully with version 7.3.1 | ~306 |
-
-### Dec 17, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #28924 | 7:29 PM | 🔵 | Plugin MCP Server Uses Bun Runtime | ~283 |
-
-### Dec 26, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32983 | 11:04 PM | 🟣 | Complete build and deployment pipeline executed | ~260 |
-
-### Jan 4, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36873 | 1:55 AM | 🔵 | Smart-Install Script Analyzed for Homebrew Path Implementation | ~466 |
-
-### Jan 7, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #38169 | 7:21 PM | 🔵 | SessionStart Hook Output Pattern Investigation Complete | ~464 |
-| #38168 | " | 🔵 | Smart-Install Script Outputs All Status Messages to stderr via console.error | ~438 |
-| #38167 | 7:20 PM | 🔵 | Context-Hook Uses stdin Event Handlers for Non-TTY JSON Output Mode | ~396 |
-| #38166 | " | 🔵 | User-Message-Hook Executes at Top Level with Await and Exit Code 1 | ~423 |
-| #38165 | " | 🔵 | Context-Hook Has Minimal Console Output in Compiled Code | ~333 |
-| #38164 | " | 🔵 | Worker-Service Script is Large 1575-Line Multi-Purpose Service Manager | ~352 |
-| #38163 | 7:19 PM | 🔵 | Worker-Service Script Uses console.log and console.error for Output | ~385 |
-| #38162 | " | 🔵 | Smart-Install Script Auto-Installs Bun and UV Dependencies | ~495 |
-| #38161 | " | 🔵 | User-Message-Hook Outputs to stderr and Exits with Code 1 | ~211 |
-| #38160 | 7:18 PM | 🔵 | Context-Hook Returns JSON with hookSpecificOutput Structure | ~470 |
-</claude-mem-context>
@@ -1,60 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Nov 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #3910 | 8:28 PM | ✅ | Refined stats counter visual design | ~343 |
-| #3909 | " | 🟣 | Added clarifying descriptions to settings UI | ~335 |
-| #3812 | 6:08 PM | 🟣 | Enhanced card typography and centered content layout | ~358 |
-
-### Nov 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #5133 | 7:29 PM | ✅ | Version 5.2.3 Released with Build Process | ~487 |
-| #4916 | 1:49 PM | ⚖️ | Claude Mem Pro Premium Offering Implementation Plan Finalized | ~946 |
-| #4902 | 1:35 PM | 🟣 | Claude Mem Pro Premium Project Initialization | ~679 |
-| #4901 | 1:31 PM | ⚖️ | Premium claude-mem Project Architecture and Planning | ~797 |
-
-### Dec 1, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #18480 | 3:39 PM | ✅ | Successfully Rebuilt Plugin After Merge Conflict Resolution | ~294 |
-
-### Dec 4, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20052 | 3:23 PM | ✅ | Built and deployed version 6.5.2 to marketplace | ~321 |
-
-### Dec 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22557 | 1:08 AM | ✅ | Build completed for version 7.0.3 | ~342 |
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23444 | 2:25 PM | 🟣 | Build Pipeline Execution Successful | ~293 |
-
-### Dec 16, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #27554 | 4:48 PM | ✅ | Project built successfully with version 7.3.1 | ~306 |
-
-### Dec 26, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32983 | 11:04 PM | 🟣 | Complete build and deployment pipeline executed | ~260 |
-| #32965 | 10:53 PM | 🔵 | Found plugin/ui/viewer.html - potential styling source | ~201 |
-| #32966 | " | 🔵 | viewer.html contains modal CSS including modal-header and modal-body | ~218 |
-| #32967 | " | 🔵 | ContextSettingsModal.tsx uses CSS classes defined in viewer.html | ~218 |
-| #32968 | " | 🔵 | Need to add CSS for footer to viewer.html | ~223 |
-</claude-mem-context>
@@ -1130,6 +1130,19 @@
      color: var(--color-text-muted);
    }

+    /* Merged-into-parent provenance badge */
+    .card-merged-badge {
+      padding: 1px 6px;
+      border-radius: 3px;
+      font-size: 9px;
+      font-weight: 500;
+      letter-spacing: 0.02em;
+      color: var(--color-text-muted);
+      background: var(--color-type-badge-bg);
+      border: 1px solid var(--color-border-primary);
+      opacity: 0.85;
+    }
+
    .summary-card {
      border-color: var(--color-border-summary);
      background: var(--color-bg-summary);
@@ -1,22 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 19, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #30153 | 8:24 PM | 🔵 | Context Builder Creates Formatted Email Investigation Context | ~384 |
-| #30152 | " | 🔵 | Ragtime Current Implementation: Manual Context Injection Via buildContextForEmail | ~357 |
-
-### Dec 20, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #30437 | 4:23 PM | 🔵 | Ragtime processes emails through Claude Agent SDK with claude-mem plugin | ~397 |
-| #30436 | 4:22 PM | 🔵 | Ragtime displays worker URL on localhost:37777 | ~219 |
-| #30340 | 3:42 PM | 🔄 | Relocated simple ragtime.ts to ragtime folder | ~219 |
-| #30339 | 3:41 PM | ✅ | Deleted overengineered ragtime.ts script | ~201 |
-| #30336 | 3:40 PM | 🔵 | Ragtime Email Corpus Processor Architecture | ~495 |
-| #30335 | " | 🔵 | Ragtime Uses Separate Noncommercial License | ~259 |
-| #30252 | 3:17 PM | 🟣 | Multi-Format Email Corpus Loader | ~436 |
-</claude-mem-context>
@@ -1 +0,0 @@
-Never read built source files in this directory. These are compiled outputs — read the source files in `src/` instead.
@@ -1,137 +0,0 @@
-# Error Handling Anti-Pattern Rules
-
-This folder contains `detect-error-handling-antipatterns.ts` - run it before committing any error handling changes.
-
-## The Try-Catch Problem That Cost 10 Hours
-
-A single overly-broad try-catch block wasted 10 hours of debugging time by silently swallowing errors.
-**This pattern is BANNED.**
-
-## BEFORE You Write Any Try-Catch
-
-**RUN THIS TEST FIRST:**
-```bash
-bun run scripts/anti-pattern-test/detect-error-handling-antipatterns.ts
-```
-
-**You MUST answer these 5 questions to the user BEFORE writing try-catch:**
-
-1. **What SPECIFIC error am I catching?** (Name the error type: `FileNotFoundError`, `NetworkTimeout`, `ValidationError`)
-2. **Show documentation proving this error can occur** (Link to docs or show me the source code)
-3. **Why can't this error be prevented?** (If it can be prevented, prevent it instead)
-4. **What will the catch block DO?** (Must include logging + either rethrow OR explicit fallback)
-5. **Why shouldn't this error propagate?** (Justify swallowing it rather than letting caller handle)
-
-**If you cannot answer ALL 5 questions with specifics, DO NOT write the try-catch.**
-
-## FORBIDDEN PATTERNS (Zero Tolerance)
-
-### CRITICAL - Never Allowed
-
-```typescript
-// FORBIDDEN: Empty catch
-try {
-  doSomething();
-} catch {}
-
-// FORBIDDEN: Catch without logging
-try {
-  doSomething();
-} catch (error) {
-  return null;  // Silent failure!
-}
-
-// FORBIDDEN: Large try blocks (>10 lines)
-try {
-  // 50 lines of code
-  // Multiple operations
-  // Different failure modes
-} catch (error) {
-  logger.error('Something failed');  // Which thing?!
-}
-
-// FORBIDDEN: Promise empty catch
-promise.catch(() => {});  // Error disappears into void
-
-// FORBIDDEN: Try-catch to fix TypeScript errors
-try {
-  // @ts-ignore
-  const value = response.propertyThatDoesntExist;
-} catch {}
-```
-
-### ALLOWED Patterns
-
-```typescript
-// GOOD: Specific, logged, explicit handling
-try {
-  await fetch(url);
-} catch (error) {
-  if (error instanceof NetworkError) {
-    logger.warn('SYNC', 'Network request failed, will retry', { url }, error);
-    return null;  // Explicit: null means "fetch failed"
-  }
-  throw error;  // Unexpected errors propagate
-}
-
-// GOOD: Minimal scope, clear recovery
-try {
-  JSON.parse(data);
-} catch (error) {
-  logger.error('CONFIG', 'Corrupt settings file, using defaults', {}, error);
-  return DEFAULT_SETTINGS;
-}
-
-// GOOD: Fire-and-forget with logging
-backgroundTask()
-  .catch(error => logger.warn('BACKGROUND', 'Task failed', {}, error));
-
-// GOOD: Ignored anti-pattern for genuine hot paths only
-try {
-  checkIfProcessAlive(pid);
-} catch (error) {
-  // [ANTI-PATTERN IGNORED]: Tight loop checking 100s of PIDs during cleanup
-  return false;
-}
-```
-
-## Ignoring Anti-Patterns (Rare)
-
-**Only for genuine hot paths** where logging would cause performance problems:
-
-```typescript
-// [ANTI-PATTERN IGNORED]: Reason why logging is impossible
-```
-
-**Rules:**
- **Hot paths only** - code in tight loops called 1000s of times
- If you can add logging, ADD LOGGING - don't ignore
- Valid examples:
-  - "Tight loop checking process exit status during cleanup"
-  - "Health check polling every 100ms"
- Invalid examples:
-  - "Expected JSON parse failures" - Just add logger.debug
-  - "Common fallback path" - Just add logger.debug
-
-## The Meta-Rule
-
-**UNCERTAINTY TRIGGERS RESEARCH, NOT TRY-CATCH**
-
-When you're unsure if a property exists or a method signature is correct:
-1. **READ** the source code or documentation
-2. **VERIFY** with the Read tool
-3. **USE** TypeScript types to catch errors at compile time
-4. **WRITE** code you KNOW is correct
-
-Never use try-catch to paper over uncertainty. That wastes hours of debugging time later.
-
-## Critical Path Protection
-
-These files are **NEVER** allowed to have catch-and-continue:
- `SDKAgent.ts` - Errors must propagate, not hide
- `GeminiAgent.ts` - Must fail loud, not silent
- `OpenRouterAgent.ts` - Must fail loud, not silent
- `SessionStore.ts` - Database errors must propagate
- `worker-service.ts` - Core service errors must be visible
-
-On critical paths, prefer **NO TRY-CATCH** and let errors propagate naturally.
@@ -0,0 +1,185 @@
+#!/usr/bin/env bun
+/**
+ * cwd-remap — Rewrite sdk_sessions.project (+ observations.project,
+ * session_summaries.project) using the cwd captured per-message in
+ * pending_messages.cwd as the single source of truth.
+ *
+ * For each distinct cwd:
+ *   - git -C <cwd> rev-parse --git-dir  AND  --git-common-dir
+ *     If they differ → worktree. parent = basename(dirname(common-dir)),
+ *     project = parent/<basename(cwd)>.
+ *     Else → project = basename(cwd).
+ *   - If the directory doesn't exist, or git errors, skip that cwd.
+ *
+ * Usage:
+ *   bun scripts/cwd-remap.ts          # dry-run (default)
+ *   bun scripts/cwd-remap.ts --apply  # write updates in a single transaction
+ */
+
+import { Database } from 'bun:sqlite';
+import { homedir } from 'os';
+import { join, basename, dirname } from 'path';
+import { existsSync, copyFileSync } from 'fs';
+import { spawnSync } from 'child_process';
+
+const DB_PATH = join(homedir(), '.claude-mem', 'claude-mem.db');
+const APPLY = process.argv.includes('--apply');
+
+type Classification =
+  | { kind: 'main'; project: string }
+  | { kind: 'worktree'; project: string; parent: string }
+  | { kind: 'skip'; reason: string };
+
+function git(cwd: string, args: string[]): string | null {
+  const r = spawnSync('git', ['-C', cwd, ...args], { encoding: 'utf8' });
+  if (r.status !== 0) {
+    const stderr = (r.stderr ?? '').trim();
+    if (stderr && !/not a git repository/i.test(stderr)) {
+      console.error(`git ${args.join(' ')} failed in ${cwd}: ${stderr}`);
+    }
+    return null;
+  }
+  return r.stdout.trim();
+}
+
+function classify(cwd: string): Classification {
+  if (!existsSync(cwd)) return { kind: 'skip', reason: 'cwd-missing' };
+
+  const gitDir = git(cwd, ['rev-parse', '--absolute-git-dir']);
+  if (!gitDir) return { kind: 'skip', reason: 'not-a-git-repo' };
+
+  const commonDir = git(cwd, ['rev-parse', '--path-format=absolute', '--git-common-dir']);
+  if (!commonDir) return { kind: 'skip', reason: 'no-common-dir' };
+
+  // Use the worktree root, not the cwd — a session may be in a subdir.
+  const toplevel = git(cwd, ['rev-parse', '--show-toplevel']);
+  if (!toplevel) return { kind: 'skip', reason: 'no-toplevel' };
+  const leaf = basename(toplevel);
+
+  if (gitDir === commonDir) {
+    return { kind: 'main', project: leaf };
+  }
+
+  // worktree: common-dir = <parent-repo>/.git (normal) or <parent>.git (bare).
+  // Normal: dirname strips the trailing /.git. Bare: strip the .git suffix.
+  const parentRepoDir = commonDir.endsWith('/.git')
+    ? dirname(commonDir)
+    : commonDir.replace(/\.git$/, '');
+  const parent = basename(parentRepoDir);
+  return { kind: 'worktree', project: `${parent}/${leaf}`, parent };
+}
+
+function main() {
+  if (!existsSync(DB_PATH)) {
+    console.error(`DB not found at ${DB_PATH}`);
+    process.exit(1);
+  }
+
+  if (APPLY) {
+    const backup = `${DB_PATH}.bak-cwd-remap-${Date.now()}`;
+    copyFileSync(DB_PATH, backup);
+    console.log(`Backup created: ${backup}`);
+  }
+
+  const db = new Database(DB_PATH);
+
+  const cwdRows = db.prepare(`
+    SELECT cwd, COUNT(*) AS messages
+    FROM pending_messages
+    WHERE cwd IS NOT NULL AND cwd != ''
+    GROUP BY cwd
+  `).all() as Array<{ cwd: string; messages: number }>;
+
+  console.log(`Classifying ${cwdRows.length} distinct cwds via git...`);
+
+  const byCwd = new Map<string, Classification>();
+  const counts = { main: 0, worktree: 0, skip: 0 };
+  for (const { cwd } of cwdRows) {
+    const c = classify(cwd);
+    byCwd.set(cwd, c);
+    counts[c.kind]++;
+  }
+  console.log(`  main=${counts.main}  worktree=${counts.worktree}  skip=${counts.skip}`);
+
+  // Skipped cwds (so user sees what's missing)
+  const skipped = [...byCwd.entries()].filter(([, c]) => c.kind === 'skip') as Array<[string, Extract<Classification, { kind: 'skip' }>]>;
+  if (skipped.length) {
+    console.log('\nSkipped cwds:');
+    for (const [cwd, c] of skipped) console.log(`  [${c.reason}] ${cwd}`);
+  }
+
+  // Per-session target: use the EARLIEST pending_messages.cwd for each session.
+  // (Dominant-cwd is wrong: claude-mem's own hooks run from nested dirs like
+  //  `.context/claude-mem/` and dominate the count, misattributing the session.)
+  const sessionRows = db.prepare(`
+    SELECT s.id AS session_id, s.memory_session_id, s.content_session_id, s.project AS old_project, p.cwd
+    FROM sdk_sessions s
+    JOIN pending_messages p ON p.content_session_id = s.content_session_id
+    WHERE p.cwd IS NOT NULL AND p.cwd != ''
+      AND p.id = (
+        SELECT MIN(p2.id) FROM pending_messages p2
+        WHERE p2.content_session_id = s.content_session_id
+          AND p2.cwd IS NOT NULL AND p2.cwd != ''
+      )
+  `).all() as Array<{ session_id: number; memory_session_id: string | null; content_session_id: string; old_project: string; cwd: string }>;
+
+  type Target = { sessionId: number; memorySessionId: string | null; contentSessionId: string; oldProject: string; newProject: string; cwd: string };
+  const perSession = new Map<number, Target>();
+
+  for (const r of sessionRows) {
+    const c = byCwd.get(r.cwd);
+    if (!c || c.kind === 'skip') continue;
+    perSession.set(r.session_id, {
+      sessionId: r.session_id,
+      memorySessionId: r.memory_session_id,
+      contentSessionId: r.content_session_id,
+      oldProject: r.old_project,
+      newProject: c.project,
+      cwd: r.cwd,
+    });
+  }
+
+  const targets = [...perSession.values()].filter(t => t.oldProject !== t.newProject);
+
+  console.log(`\nSessions linked to a classified cwd: ${perSession.size}`);
+  console.log(`Sessions whose project would change: ${targets.length}`);
+
+  const summary = new Map<string, number>();
+  for (const t of targets) {
+    const key = `${t.oldProject}  →  ${t.newProject}`;
+    summary.set(key, (summary.get(key) ?? 0) + 1);
+  }
+  const rows = [...summary.entries()]
+    .map(([mapping, n]) => ({ mapping, sessions: n }))
+    .sort((a, b) => b.sessions - a.sessions);
+  console.log('\nTop mappings:');
+  console.table(rows.slice(0, 30));
+  if (rows.length > 30) console.log(`  …and ${rows.length - 30} more mappings`);
+
+  if (!APPLY) {
+    console.log('\nDry-run only. Re-run with --apply to perform UPDATEs.');
+    db.close();
+    return;
+  }
+
+  const updSession = db.prepare('UPDATE sdk_sessions      SET project = ? WHERE id = ?');
+  const updObs     = db.prepare('UPDATE observations      SET project = ? WHERE memory_session_id = ?');
+  const updSum     = db.prepare('UPDATE session_summaries SET project = ? WHERE memory_session_id = ?');
+
+  let sessionN = 0, obsN = 0, sumN = 0;
+  const tx = db.transaction(() => {
+    for (const t of targets) {
+      sessionN += updSession.run(t.newProject, t.sessionId).changes;
+      if (t.memorySessionId) {
+        obsN += updObs.run(t.newProject, t.memorySessionId).changes;
+        sumN += updSum.run(t.newProject, t.memorySessionId).changes;
+      }
+    }
+  });
+  tx();
+
+  console.log(`\nApplied. sessions=${sessionN} observations=${obsN} session_summaries=${sumN}`);
+  db.close();
+}
+
+main();
@@ -1,3 +0,0 @@
-<claude-mem-context>
-
-</claude-mem-context>
@@ -1,34 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23825 | 11:12 PM | ✅ | Worker Port Set to 38888 for Migration Phase | ~283 |
-| #23824 | " | 🔵 | Worker Port Sourced from getWorkerPort() Utility | ~247 |
-| #23816 | 10:52 PM | 🟣 | Worker CLI Command Interface Created | ~325 |
-
-### Dec 11, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24060 | 2:58 PM | 🔴 | Worker CLI Start Command Exit Behavior Fixed | ~232 |
-
-### Dec 12, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24359 | 7:00 PM | 🟣 | Phase 1 Critical Code Fixes Completed via Agent Task | ~441 |
-| #24358 | 6:59 PM | ✅ | Completed Phase 1 Code Fixes for better-sqlite3 Migration | ~385 |
-| #24348 | 6:57 PM | 🔴 | Added Defensive Break Statement to worker-cli.ts Restart Case | ~269 |
-| #24345 | " | 🔵 | worker-cli.ts Missing Break Statement in Switch Case | ~318 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26766 | 11:30 PM | ⚖️ | Root Cause Identified: Missing Post-Install Worker Restart Trigger in Plugin Update Flow | ~604 |
-| #26722 | 11:23 PM | 🔵 | Worker CLI TypeScript Source Shows Simple ProcessManager Delegation | ~394 |
-| #26721 | " | 🔵 | Worker CLI Source Code Shows Simple Restart Logic Without Delays | ~425 |
-</claude-mem-context>
@@ -1,3 +0,0 @@
-<claude-mem-context>
-
-</claude-mem-context>
@@ -1,3 +0,0 @@
-<claude-mem-context>
-
-</claude-mem-context>
@@ -101,6 +101,44 @@ export function runStatusCommand(): void {
  spawnBunWorkerCommand('status');
 }

+/**
+ * Stamp merged-worktree provenance on observations/summaries and keep Chroma
+ * metadata in lockstep. Delegates to the worker-service.cjs `adopt` subcommand
+ * so adoption runs in Bun (needed for bun:sqlite) while preserving the user's
+ * working directory — that's what the engine uses to locate the parent repo.
+ */
+export function runAdoptCommand(extraArgs: string[] = []): void {
+  ensureInstalledOrExit();
+  const bunPath = resolveBunOrExit();
+  const workerScript = workerServiceScriptPath();
+
+  if (!existsSync(workerScript)) {
+    console.error(pc.red(`Worker script not found at: ${workerScript}`));
+    console.error('The installation may be corrupted. Try: npx claude-mem install');
+    process.exit(1);
+  }
+
+  // Pass user's cwd explicitly via --cwd because we override cwd on spawn to
+  // marketplaceDirectory() (required for the worker's own file resolution).
+  const userCwd = process.cwd();
+  const args = [workerScript, 'adopt', '--cwd', userCwd, ...extraArgs];
+
+  const child = spawn(bunPath, args, {
+    stdio: 'inherit',
+    cwd: marketplaceDirectory(),
+    env: process.env,
+  });
+
+  child.on('error', (error) => {
+    console.error(pc.red(`Failed to start Bun: ${error.message}`));
+    process.exit(1);
+  });
+
+  child.on('close', (exitCode) => {
+    process.exit(exitCode ?? 0);
+  });
+}
+
 /**
 * Search the worker API at `GET /api/search?query=<query>`.
 */
@@ -52,6 +52,7 @@ ${pc.bold('Runtime Commands')} (requires Bun, delegates to installed plugin):
  ${pc.cyan('npx claude-mem restart')}              Restart worker service
  ${pc.cyan('npx claude-mem status')}               Show worker status
  ${pc.cyan('npx claude-mem search <query>')}       Search observations
+  ${pc.cyan('npx claude-mem adopt [--dry-run] [--branch <name>]')}    Stamp merged worktrees into parent project
  ${pc.cyan('npx claude-mem transcript watch')}     Start transcript watcher

 ${pc.bold('IDE Identifiers')}:
@@ -145,6 +146,13 @@ async function main(): Promise<void> {
      break;
    }

+    // -- Adopt merged worktrees -------------------------------------------
+    case 'adopt': {
+      const { runAdoptCommand } = await import('./commands/runtime.js');
+      runAdoptCommand(args.slice(1));
+      break;
+    }
+
    // -- Transcript --------------------------------------------------------
    case 'transcript': {
      const subCommand = args[1]?.toLowerCase();
@@ -1,61 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23832 | 11:15 PM | 🔵 | Current worker-service.ts Lacks Admin Endpoints | ~393 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26740 | 11:26 PM | 🔵 | Worker Service Refactored to Orchestrator with Background Initialization | ~421 |
-| #26739 | 11:25 PM | 🔵 | Worker Service Architecture Uses Domain Services and Background Initialization | ~438 |
-| #26255 | 8:31 PM | 🔵 | Context Generator Timeline Rendering Logic Details File Grouping Implementation | ~397 |
-| #26251 | 8:30 PM | 🔵 | Worker Service Orchestrates Domain Services and Route Handlers | ~292 |
-| #26246 | 8:29 PM | 🔵 | Context Generator Implements Rich Date-Grouped Timeline Format | ~468 |
-
-### Dec 17, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #28548 | 4:49 PM | 🔵 | Worker service cleanup method uses Unix-specific process management | ~323 |
-| #28446 | 4:23 PM | 🔵 | Worker Service Refactored to Orchestrator Pattern | ~529 |
-
-### Dec 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #29340 | 3:11 PM | ✅ | Constructor Initialization Comment Updated | ~267 |
-| #29339 | " | ✅ | Class Member Comment Updated in WorkerService | ~267 |
-| #29338 | " | ✅ | Service Import Comment Updated | ~222 |
-| #29337 | 3:10 PM | ✅ | Terminology Update in Worker Service Documentation | ~268 |
-| #29239 | 12:11 AM | 🔵 | Worker Service Refactored as Domain-Driven Orchestrator | ~477 |
-
-### Dec 20, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #30808 | 6:05 PM | 🔴 | Fixed worker readiness check to fail on initialization errors | ~315 |
-| #30800 | 6:03 PM | 🔵 | Dual Error Logging in Background Initialization | ~367 |
-| #30799 | " | 🔵 | Background Initialization Invocation Pattern | ~365 |
-| #30797 | " | 🔵 | Background Initialization Sequence and Error Handler Confirmed | ~450 |
-| #30795 | 6:02 PM | 🔵 | Readiness Endpoint Returns 503 During Initialization | ~397 |
-| #30793 | " | 🔵 | Dual Initialization State Tracking Pattern | ~388 |
-| #30791 | " | 🔵 | Worker Service Constructor Defers SearchRoutes Initialization | ~387 |
-| #30790 | " | 🔵 | Initialization Promise Resolver Pattern Located | ~321 |
-| #30788 | " | 🔵 | Worker Service Initialization Resolves Promise Despite Errors | ~388 |
-
-### Jan 1, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #35654 | 11:29 PM | ✅ | Added APPROVED OVERRIDE annotation for instruction loading HTTP route error handler | ~339 |
-| #35651 | 11:28 PM | ✅ | Added APPROVED OVERRIDE annotation for shutdown error handler with process.exit | ~354 |
-| #35649 | " | ✅ | Added APPROVED OVERRIDE annotation for readiness check retry loop error handling | ~374 |
-| #35647 | " | ✅ | Added APPROVED OVERRIDE annotation for port availability probe error handling | ~327 |
-| #35646 | " | ✅ | Added APPROVED OVERRIDE annotation for Cursor context file update error handling | ~342 |
-| #35643 | 11:27 PM | ✅ | Added APPROVED OVERRIDE annotation for PID file cleanup error handling | ~320 |
-</claude-mem-context>
@@ -130,11 +130,14 @@ export async function generateContext(
  const config = loadContextConfig();
  const cwd = input?.cwd ?? process.cwd();
  const context = getProjectContext(cwd);
-  const project = context.primary;
  const platformSource = input?.platform_source;

-  // Use provided projects array (for worktree support) or fall back to all known projects
-  const projects = input?.projects ?? context.allProjects;
+  // Single source of truth: explicit projects override cwd-derived context.
+  // `project` (used for header + single-project query) is always the last entry
+  // of `projects` so the empty-state header and the query target stay in sync
+  // when a caller passes `projects` without a matching cwd (e.g. worker route).
+  const projects = input?.projects?.length ? input.projects : context.allProjects;
+  const project = projects[projects.length - 1] ?? context.primary;

  // Full mode: fetch all observations but keep normal rendering (level 1 summaries)
  if (input?.full) {
@@ -52,7 +52,7 @@ export function queryObservations(
      o.created_at_epoch
    FROM observations o
    LEFT JOIN sdk_sessions s ON o.memory_session_id = s.memory_session_id
-    WHERE o.project = ?
+    WHERE (o.project = ? OR o.merged_into_project = ?)
      AND type IN (${typePlaceholders})
      AND EXISTS (
        SELECT 1 FROM json_each(o.concepts)
@@ -62,6 +62,7 @@ export function queryObservations(
    ORDER BY o.created_at_epoch DESC
    LIMIT ?
  `).all(
+    project,
    project,
    ...typeArray,
    ...conceptArray,
@@ -93,12 +94,12 @@ export function querySummaries(
      ss.created_at_epoch
    FROM session_summaries ss
    LEFT JOIN sdk_sessions s ON ss.memory_session_id = s.memory_session_id
-    WHERE ss.project = ?
+    WHERE (ss.project = ? OR ss.merged_into_project = ?)
      ${platformSource ? "AND COALESCE(s.platform_source, 'claude') = ?" : ''}
    ORDER BY ss.created_at_epoch DESC
    LIMIT ?
  `).all(
-    ...[project, ...(platformSource ? [platformSource] : []), config.sessionCount + SUMMARY_LOOKAHEAD]
+    ...[project, project, ...(platformSource ? [platformSource] : []), config.sessionCount + SUMMARY_LOOKAHEAD]
  ) as SessionSummary[];
 }

@@ -141,7 +142,8 @@ export function queryObservationsMulti(
      o.project
    FROM observations o
    LEFT JOIN sdk_sessions s ON o.memory_session_id = s.memory_session_id
-    WHERE o.project IN (${projectPlaceholders})
+    WHERE (o.project IN (${projectPlaceholders})
+           OR o.merged_into_project IN (${projectPlaceholders}))
      AND type IN (${typePlaceholders})
      AND EXISTS (
        SELECT 1 FROM json_each(o.concepts)
@@ -151,6 +153,7 @@ export function queryObservationsMulti(
    ORDER BY o.created_at_epoch DESC
    LIMIT ?
  `).all(
+    ...projects,
    ...projects,
    ...typeArray,
    ...conceptArray,
@@ -189,11 +192,12 @@ export function querySummariesMulti(
      ss.project
    FROM session_summaries ss
    LEFT JOIN sdk_sessions s ON ss.memory_session_id = s.memory_session_id
-    WHERE ss.project IN (${projectPlaceholders})
+    WHERE (ss.project IN (${projectPlaceholders})
+           OR ss.merged_into_project IN (${projectPlaceholders}))
      ${platformSource ? "AND COALESCE(s.platform_source, 'claude') = ?" : ''}
    ORDER BY ss.created_at_epoch DESC
    LIMIT ?
-  `).all(...projects, ...(platformSource ? [platformSource] : []), config.sessionCount + SUMMARY_LOOKAHEAD) as SessionSummary[];
+  `).all(...projects, ...projects, ...(platformSource ? [platformSource] : []), config.sessionCount + SUMMARY_LOOKAHEAD) as SessionSummary[];
 }

 /**
@@ -35,7 +35,7 @@ function formatHeaderDateTime(): string {
 */
 export function renderAgentHeader(project: string): string[] {
  return [
-    `# $CMEM ${project} ${formatHeaderDateTime()}`,
+    `# [${project}] recent context, ${formatHeaderDateTime()}`,
    ''
  ];
 }
@@ -223,5 +223,5 @@ export function renderAgentFooter(totalDiscoveryTokens: number, totalReadTokens:
 * Render agent empty state
 */
 export function renderAgentEmptyState(project: string): string {
-  return `# $CMEM ${project} ${formatHeaderDateTime()}\n\nNo previous sessions found.`;
+  return `# [${project}] recent context, ${formatHeaderDateTime()}\n\nNo previous sessions found.`;
 }
@@ -1,12 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-<!-- This section is auto-generated by claude-mem. Edit content outside the tags. -->
-
-### Jan 25, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #41877 | 12:09 PM | ⚖️ | Deploy Existing Consumer Preview Without Creating New Packages | ~361 |
-| #41873 | 12:03 PM | 🔵 | Claude-mem mode configuration system types documented | ~504 |
-</claude-mem-context>
@@ -1,10 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Jan 4, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36864 | 1:52 AM | 🔵 | ProcessManager Module Imports Reviewed | ~245 |
-| #36860 | 1:50 AM | 🔵 | ProcessManager Source Code Reviewed for WMIC Implementation | ~608 |
-</claude-mem-context>
@@ -10,8 +10,8 @@

 import path from 'path';
 import { homedir } from 'os';
-import { existsSync, writeFileSync, readFileSync, unlinkSync, mkdirSync, rmSync, statSync, utimesSync } from 'fs';
-import { exec, execSync, spawn } from 'child_process';
+import { existsSync, writeFileSync, readFileSync, unlinkSync, mkdirSync, rmSync, statSync, utimesSync, copyFileSync } from 'fs';
+import { exec, execSync, spawn, spawnSync } from 'child_process';
 import { promisify } from 'util';
 import { logger } from '../../utils/logger.js';
 import { HOOK_TIMEOUTS } from '../../shared/hook-constants.js';
@@ -677,6 +677,161 @@ export function runOneTimeChromaMigration(dataDirectory?: string): void {
  logger.info('SYSTEM', 'Chroma migration marker written', { markerPath });
 }

+const CWD_REMAP_MARKER_FILENAME = '.cwd-remap-applied-v1';
+
+type CwdClassification =
+  | { kind: 'main'; project: string }
+  | { kind: 'worktree'; project: string }
+  | { kind: 'skip' };
+
+function gitQuery(cwd: string, args: string[]): string | null {
+  const r = spawnSync('git', ['-C', cwd, ...args], {
+    encoding: 'utf8',
+    timeout: 5000
+  });
+  if (r.status !== 0) return null;
+  return (r.stdout ?? '').trim();
+}
+
+function classifyCwdForRemap(cwd: string): CwdClassification {
+  if (!existsSync(cwd)) return { kind: 'skip' };
+
+  const gitDir = gitQuery(cwd, ['rev-parse', '--absolute-git-dir']);
+  if (!gitDir) return { kind: 'skip' };
+
+  const commonDir = gitQuery(cwd, ['rev-parse', '--path-format=absolute', '--git-common-dir']);
+  if (!commonDir) return { kind: 'skip' };
+
+  const toplevel = gitQuery(cwd, ['rev-parse', '--show-toplevel']);
+  if (!toplevel) return { kind: 'skip' };
+  const leaf = path.basename(toplevel);
+
+  if (gitDir === commonDir) {
+    return { kind: 'main', project: leaf };
+  }
+
+  const parentRepoDir = commonDir.endsWith('/.git')
+    ? path.dirname(commonDir)
+    : commonDir.replace(/\.git$/, '');
+  const parent = path.basename(parentRepoDir);
+  return { kind: 'worktree', project: `${parent}/${leaf}` };
+}
+
+/**
+ * One-time remap of sdk_sessions.project (+ observations.project,
+ * session_summaries.project) using the cwd captured in pending_messages.cwd
+ * as the source of truth. Required because pre-worktree builds stored bare
+ * project names that collide across parent/worktree checkouts.
+ *
+ * Backs up the DB before writes. Idempotent via marker file. Skips silently
+ * if the DB or pending_messages table doesn't exist yet (fresh install).
+ *
+ * @param dataDirectory - Override for DATA_DIR (used in tests)
+ */
+export function runOneTimeCwdRemap(dataDirectory?: string): void {
+  const effectiveDataDir = dataDirectory ?? DATA_DIR;
+  const markerPath = path.join(effectiveDataDir, CWD_REMAP_MARKER_FILENAME);
+  const dbPath = path.join(effectiveDataDir, 'claude-mem.db');
+
+  if (existsSync(markerPath)) {
+    logger.debug('SYSTEM', 'cwd-remap marker exists, skipping');
+    return;
+  }
+
+  if (!existsSync(dbPath)) {
+    mkdirSync(effectiveDataDir, { recursive: true });
+    writeFileSync(markerPath, new Date().toISOString());
+    logger.debug('SYSTEM', 'No DB present, cwd-remap marker written without work', { dbPath });
+    return;
+  }
+
+  logger.warn('SYSTEM', 'Running one-time cwd-based project remap', { dbPath });
+
+  let db: import('bun:sqlite').Database | null = null;
+  try {
+    const { Database } = require('bun:sqlite') as typeof import('bun:sqlite');
+
+    const probe = new Database(dbPath, { readonly: true });
+    const hasPending = probe.prepare(
+      "SELECT name FROM sqlite_master WHERE type='table' AND name='pending_messages'"
+    ).get() as { name: string } | undefined;
+    probe.close();
+
+    if (!hasPending) {
+      mkdirSync(effectiveDataDir, { recursive: true });
+      writeFileSync(markerPath, new Date().toISOString());
+      logger.info('SYSTEM', 'pending_messages table not present, cwd-remap skipped');
+      return;
+    }
+
+    const backup = `${dbPath}.bak-cwd-remap-${Date.now()}`;
+    copyFileSync(dbPath, backup);
+    logger.info('SYSTEM', 'DB backed up before cwd-remap', { backup });
+
+    db = new Database(dbPath);
+
+    const cwdRows = db.prepare(`
+      SELECT cwd FROM pending_messages
+      WHERE cwd IS NOT NULL AND cwd != ''
+      GROUP BY cwd
+    `).all() as Array<{ cwd: string }>;
+
+    const byCwd = new Map<string, CwdClassification>();
+    for (const { cwd } of cwdRows) byCwd.set(cwd, classifyCwdForRemap(cwd));
+
+    const sessionRows = db.prepare(`
+      SELECT s.id AS session_id, s.memory_session_id, s.project AS old_project, p.cwd
+      FROM sdk_sessions s
+      JOIN pending_messages p ON p.content_session_id = s.content_session_id
+      WHERE p.cwd IS NOT NULL AND p.cwd != ''
+        AND p.id = (
+          SELECT MIN(p2.id) FROM pending_messages p2
+          WHERE p2.content_session_id = s.content_session_id
+            AND p2.cwd IS NOT NULL AND p2.cwd != ''
+        )
+    `).all() as Array<{ session_id: number; memory_session_id: string | null; old_project: string; cwd: string }>;
+
+    type Target = { sessionId: number; memorySessionId: string | null; newProject: string };
+    const targets: Target[] = [];
+    for (const r of sessionRows) {
+      const c = byCwd.get(r.cwd);
+      if (!c || c.kind === 'skip') continue;
+      if (r.old_project === c.project) continue;
+      targets.push({ sessionId: r.session_id, memorySessionId: r.memory_session_id, newProject: c.project });
+    }
+
+    if (targets.length === 0) {
+      logger.info('SYSTEM', 'cwd-remap: no sessions need updating');
+    } else {
+      const updSession = db.prepare('UPDATE sdk_sessions      SET project = ? WHERE id = ?');
+      const updObs     = db.prepare('UPDATE observations      SET project = ? WHERE memory_session_id = ?');
+      const updSum     = db.prepare('UPDATE session_summaries SET project = ? WHERE memory_session_id = ?');
+
+      let sessionN = 0, obsN = 0, sumN = 0;
+      const tx = db.transaction(() => {
+        for (const t of targets) {
+          sessionN += updSession.run(t.newProject, t.sessionId).changes;
+          if (t.memorySessionId) {
+            obsN += updObs.run(t.newProject, t.memorySessionId).changes;
+            sumN += updSum.run(t.newProject, t.memorySessionId).changes;
+          }
+        }
+      });
+      tx();
+
+      logger.info('SYSTEM', 'cwd-remap applied', { sessions: sessionN, observations: obsN, summaries: sumN, backup });
+    }
+
+    mkdirSync(effectiveDataDir, { recursive: true });
+    writeFileSync(markerPath, new Date().toISOString());
+    logger.info('SYSTEM', 'cwd-remap marker written', { markerPath });
+  } catch (err) {
+    logger.error('SYSTEM', 'cwd-remap failed, marker not written (will retry on next startup)', {}, err as Error);
+  } finally {
+    db?.close();
+  }
+}
+
 /**
 * Spawn a detached daemon process
 * Returns the child PID or undefined if spawn failed
@@ -0,0 +1,414 @@
+/**
+ * WorktreeAdoption - Stamp observations from merged worktrees into their parent project.
+ *
+ * Given a parent repo path, this engine:
+ *   1. Uses git to enumerate worktrees of the parent repo.
+ *   2. Classifies each worktree's branch as "merged" (in `git branch --merged HEAD`)
+ *      or manually overridden via `onlyBranch` (for squash-merge detection).
+ *   3. Stamps `merged_into_project` on `observations` and `session_summaries` rows
+ *      whose `project` matches the composite `parent/worktree` name.
+ *   4. Propagates the same metadata to Chroma so semantic search includes the
+ *      adopted rows under the parent project.
+ *
+ * `project` is never overwritten — it remains immutable provenance. The
+ * `merged_into_project` column is a virtual pointer that query layers OR into
+ * their WHERE predicates.
+ *
+ * DB lifecycle mirrors `runOneTimeCwdRemap` in ProcessManager.ts: we manage our
+ * own Database handle (open -> transaction -> close in finally) so this engine
+ * can be called on worker startup before `dbManager.initialize()` without
+ * contending on the shared handle.
+ */
+
+import path from 'path';
+import { homedir } from 'os';
+import { existsSync } from 'fs';
+import { spawnSync } from 'child_process';
+import { logger } from '../../utils/logger.js';
+import { getProjectContext } from '../../utils/project-name.js';
+import { ChromaSync } from '../sync/ChromaSync.js';
+
+const DEFAULT_DATA_DIR = path.join(homedir(), '.claude-mem');
+
+export interface AdoptionResult {
+  repoPath: string;
+  parentProject: string;
+  scannedWorktrees: number;
+  mergedBranches: string[];
+  adoptedObservations: number;
+  adoptedSummaries: number;
+  chromaUpdates: number;
+  chromaFailed: number;
+  dryRun: boolean;
+  errors: Array<{ worktree: string; error: string }>;
+}
+
+interface WorktreeEntry {
+  path: string;
+  branch: string | null;
+}
+
+const GIT_TIMEOUT_MS = 5000;
+
+class DryRunRollback extends Error {
+  constructor() {
+    super('dry-run rollback');
+    this.name = 'DryRunRollback';
+  }
+}
+
+function gitCapture(cwd: string, args: string[]): string | null {
+  const r = spawnSync('git', ['-C', cwd, ...args], {
+    encoding: 'utf8',
+    timeout: GIT_TIMEOUT_MS
+  });
+  if (r.status !== 0) return null;
+  return (r.stdout ?? '').trim();
+}
+
+/**
+ * Resolve the main working-tree root for an arbitrary cwd inside a repo or worktree.
+ * Mirrors the handling in `scripts/cwd-remap.ts:48-51`.
+ */
+function resolveMainRepoPath(cwd: string): string | null {
+  const commonDir = gitCapture(cwd, [
+    'rev-parse',
+    '--path-format=absolute',
+    '--git-common-dir'
+  ]);
+  if (!commonDir) return null;
+
+  // Normal: common-dir is "<repo>/.git". Bare: strip the trailing ".git".
+  const mainRoot = commonDir.endsWith('/.git')
+    ? path.dirname(commonDir)
+    : commonDir.replace(/\.git$/, '');
+  return existsSync(mainRoot) ? mainRoot : null;
+}
+
+function listWorktrees(mainRepo: string): WorktreeEntry[] {
+  const raw = gitCapture(mainRepo, ['worktree', 'list', '--porcelain']);
+  if (!raw) return [];
+
+  const entries: WorktreeEntry[] = [];
+  let current: Partial<WorktreeEntry> = {};
+  for (const line of raw.split('\n')) {
+    if (line.startsWith('worktree ')) {
+      if (current.path) entries.push({ path: current.path, branch: current.branch ?? null });
+      current = { path: line.slice('worktree '.length).trim(), branch: null };
+    } else if (line.startsWith('branch ')) {
+      // `branch refs/heads/<name>` — strip the ref prefix.
+      const refName = line.slice('branch '.length).trim();
+      current.branch = refName.startsWith('refs/heads/')
+        ? refName.slice('refs/heads/'.length)
+        : refName;
+    } else if (line === '' && current.path) {
+      entries.push({ path: current.path, branch: current.branch ?? null });
+      current = {};
+    }
+  }
+  if (current.path) entries.push({ path: current.path, branch: current.branch ?? null });
+  return entries;
+}
+
+function listMergedBranches(mainRepo: string): Set<string> {
+  const raw = gitCapture(mainRepo, [
+    'branch',
+    '--merged',
+    'HEAD',
+    '--format=%(refname:short)'
+  ]);
+  if (!raw) return new Set();
+  return new Set(
+    raw.split('\n').map(b => b.trim()).filter(b => b.length > 0)
+  );
+}
+
+/**
+ * Stamp `merged_into_project` on observations and session_summaries for every
+ * worktree of `opts.repoPath` whose branch has been merged into the parent's HEAD.
+ *
+ * SQL writes are idempotent: an UPDATE only touches rows where
+ * `merged_into_project IS NULL`. `result.adoptedObservations` / `adoptedSummaries`
+ * reflect the actual SQL changes on each run.
+ *
+ * Chroma patches are self-healing: the Chroma id set is built from ALL
+ * observations whose `project` matches a merged worktree (both unadopted rows
+ * AND rows previously stamped to this parent), and `updateMergedIntoProject`
+ * is idempotent, so a transient Chroma failure on an earlier run is retried
+ * automatically on the next adoption pass. `result.chromaUpdates` therefore
+ * counts the total Chroma writes performed this pass (which may exceed
+ * `adoptedObservations` when retries happen).
+ */
+export async function adoptMergedWorktrees(opts: {
+  repoPath?: string;
+  dataDirectory?: string;
+  dryRun?: boolean;
+  onlyBranch?: string;
+} = {}): Promise<AdoptionResult> {
+  const dataDirectory = opts.dataDirectory ?? DEFAULT_DATA_DIR;
+  const dryRun = opts.dryRun ?? false;
+  const startCwd = opts.repoPath ?? process.cwd();
+
+  const mainRepo = resolveMainRepoPath(startCwd);
+  const parentProject = mainRepo ? getProjectContext(mainRepo).primary : '';
+
+  const result: AdoptionResult = {
+    repoPath: mainRepo ?? startCwd,
+    parentProject,
+    scannedWorktrees: 0,
+    mergedBranches: [],
+    adoptedObservations: 0,
+    adoptedSummaries: 0,
+    chromaUpdates: 0,
+    chromaFailed: 0,
+    dryRun,
+    errors: []
+  };
+
+  if (!mainRepo) {
+    logger.debug('SYSTEM', 'Worktree adoption skipped (not a git repo)', { startCwd });
+    return result;
+  }
+
+  const dbPath = path.join(dataDirectory, 'claude-mem.db');
+  if (!existsSync(dbPath)) {
+    logger.debug('SYSTEM', 'Worktree adoption skipped (no DB yet)', { dbPath });
+    return result;
+  }
+
+  const allWorktrees = listWorktrees(mainRepo);
+  const childWorktrees = allWorktrees.filter(w => w.path !== mainRepo);
+  result.scannedWorktrees = childWorktrees.length;
+
+  if (childWorktrees.length === 0) {
+    return result;
+  }
+
+  let targets: WorktreeEntry[];
+  if (opts.onlyBranch) {
+    targets = childWorktrees.filter(w => w.branch === opts.onlyBranch);
+  } else {
+    const merged = listMergedBranches(mainRepo);
+    targets = childWorktrees.filter(w => w.branch !== null && merged.has(w.branch));
+  }
+
+  result.mergedBranches = targets
+    .map(t => t.branch)
+    .filter((b): b is string => b !== null);
+
+  if (targets.length === 0) {
+    return result;
+  }
+
+  const adoptedSqliteIds: number[] = [];
+
+  let db: import('bun:sqlite').Database | null = null;
+  try {
+    const { Database } = require('bun:sqlite') as typeof import('bun:sqlite');
+    db = new Database(dbPath);
+
+    // Schema guard: adoption may be invoked on worker startup before
+    // DatabaseManager runs migrations. If the `merged_into_project` column
+    // isn't present yet, prepared statements below will fail with
+    // "no such column", silently skipping adoption until the next restart.
+    // Return early so the next boot (post-migration) picks this up.
+    interface ColumnInfo { name: string }
+    const obsColumns = db
+      .prepare('PRAGMA table_info(observations)')
+      .all() as ColumnInfo[];
+    const sumColumns = db
+      .prepare('PRAGMA table_info(session_summaries)')
+      .all() as ColumnInfo[];
+    const obsHasColumn = obsColumns.some(c => c.name === 'merged_into_project');
+    const sumHasColumn = sumColumns.some(c => c.name === 'merged_into_project');
+    if (!obsHasColumn || !sumHasColumn) {
+      logger.debug(
+        'SYSTEM',
+        'Worktree adoption skipped (merged_into_project column missing; will run after migration)',
+        { obsHasColumn, sumHasColumn }
+      );
+      return result;
+    }
+
+    // Select ALL observations for the worktree project (both unadopted rows
+    // AND rows already stamped to this parent), not just unadopted ones. This
+    // ensures a transient Chroma failure on a prior run gets retried the next
+    // time adoption executes: SQL may already be stamped, but we re-include
+    // those ids in the Chroma patch set (updateMergedIntoProject is idempotent
+    // — it replays the same metadata write).
+    const selectObsForPatch = db.prepare(
+      `SELECT id FROM observations
+       WHERE project = ?
+         AND (merged_into_project IS NULL OR merged_into_project = ?)`
+    );
+    const updateObs = db.prepare(
+      'UPDATE observations SET merged_into_project = ? WHERE project = ? AND merged_into_project IS NULL'
+    );
+    const updateSum = db.prepare(
+      'UPDATE session_summaries SET merged_into_project = ? WHERE project = ? AND merged_into_project IS NULL'
+    );
+
+    const tx = db.transaction(() => {
+      for (const wt of targets) {
+        try {
+          const worktreeProject = getProjectContext(wt.path).primary;
+          const rows = selectObsForPatch.all(
+            worktreeProject,
+            parentProject
+          ) as Array<{ id: number }>;
+          for (const r of rows) adoptedSqliteIds.push(r.id);
+
+          // updateObs/updateSum only touch WHERE merged_into_project IS NULL,
+          // so .changes reflects only newly-adopted rows (not the re-patched ones).
+          const obsChanges = updateObs.run(parentProject, worktreeProject).changes;
+          const sumChanges = updateSum.run(parentProject, worktreeProject).changes;
+          result.adoptedObservations += obsChanges;
+          result.adoptedSummaries += sumChanges;
+        } catch (err) {
+          const message = err instanceof Error ? err.message : String(err);
+          logger.warn('SYSTEM', 'Worktree adoption skipped branch', {
+            worktree: wt.path,
+            branch: wt.branch,
+            error: message
+          });
+          result.errors.push({ worktree: wt.path, error: message });
+        }
+      }
+      if (dryRun) {
+        // Throw a dedicated error to force rollback. Caught below by instanceof check.
+        throw new DryRunRollback();
+      }
+    });
+
+    try {
+      tx();
+    } catch (err) {
+      if (err instanceof DryRunRollback) {
+        // Rolled back as intended for dry-run — counts are still useful.
+      } else {
+        throw err;
+      }
+    }
+  } finally {
+    db?.close();
+  }
+
+  if (!dryRun && adoptedSqliteIds.length > 0) {
+    const chromaSync = new ChromaSync('claude-mem');
+    try {
+      await chromaSync.updateMergedIntoProject(adoptedSqliteIds, parentProject);
+      result.chromaUpdates = adoptedSqliteIds.length;
+    } catch (err) {
+      logger.error(
+        'CHROMA_SYNC',
+        'Worktree adoption Chroma patch failed (SQL already committed)',
+        { parentProject, sqliteIdCount: adoptedSqliteIds.length },
+        err as Error
+      );
+      result.chromaFailed = adoptedSqliteIds.length;
+    } finally {
+      await chromaSync.close();
+    }
+  }
+
+  if (
+    result.adoptedObservations > 0 ||
+    result.adoptedSummaries > 0 ||
+    result.chromaUpdates > 0 ||
+    result.errors.length > 0
+  ) {
+    logger.info('SYSTEM', 'Worktree adoption applied', {
+      parentProject,
+      dryRun,
+      scannedWorktrees: result.scannedWorktrees,
+      mergedBranches: result.mergedBranches,
+      adoptedObservations: result.adoptedObservations,
+      adoptedSummaries: result.adoptedSummaries,
+      chromaUpdates: result.chromaUpdates,
+      chromaFailed: result.chromaFailed,
+      errors: result.errors.length
+    });
+  }
+
+  return result;
+}
+
+/**
+ * Run adoption once per distinct parent repo referenced by recorded cwds.
+ *
+ * Worker startup adoption cannot use `process.cwd()` as a seed — the daemon is
+ * spawned with cwd=marketplace-plugin-dir, which isn't a git repo. Instead, we
+ * derive candidate parent repos from `pending_messages.cwd` (the user's actual
+ * working directories), dedupe via `resolveMainRepoPath`, and run adoption
+ * against each. Failures on individual repos are logged but don't short-circuit
+ * the others.
+ *
+ * Safe to call before `dbManager.initialize()`: opens its own short-lived DB
+ * handle (readonly) to enumerate cwds, then delegates to `adoptMergedWorktrees`
+ * which opens its own writable handle.
+ */
+export async function adoptMergedWorktreesForAllKnownRepos(opts: {
+  dataDirectory?: string;
+  dryRun?: boolean;
+} = {}): Promise<AdoptionResult[]> {
+  const dataDirectory = opts.dataDirectory ?? DEFAULT_DATA_DIR;
+  const dbPath = path.join(dataDirectory, 'claude-mem.db');
+  const results: AdoptionResult[] = [];
+
+  if (!existsSync(dbPath)) {
+    logger.debug('SYSTEM', 'Worktree adoption skipped (no DB yet)', { dbPath });
+    return results;
+  }
+
+  const uniqueParents = new Set<string>();
+  let db: import('bun:sqlite').Database | null = null;
+  try {
+    const { Database } = require('bun:sqlite') as typeof import('bun:sqlite');
+    db = new Database(dbPath, { readonly: true });
+
+    const hasPending = db.prepare(
+      "SELECT name FROM sqlite_master WHERE type='table' AND name='pending_messages'"
+    ).get() as { name: string } | undefined;
+    if (!hasPending) {
+      logger.debug('SYSTEM', 'Worktree adoption skipped (pending_messages table missing)');
+      return results;
+    }
+
+    const cwdRows = db.prepare(`
+      SELECT cwd FROM pending_messages
+      WHERE cwd IS NOT NULL AND cwd != ''
+      GROUP BY cwd
+    `).all() as Array<{ cwd: string }>;
+
+    for (const { cwd } of cwdRows) {
+      const mainRepo = resolveMainRepoPath(cwd);
+      if (mainRepo) uniqueParents.add(mainRepo);
+    }
+  } finally {
+    db?.close();
+  }
+
+  if (uniqueParents.size === 0) {
+    logger.debug('SYSTEM', 'Worktree adoption found no known parent repos');
+    return results;
+  }
+
+  for (const repoPath of uniqueParents) {
+    try {
+      const result = await adoptMergedWorktrees({
+        repoPath,
+        dataDirectory,
+        dryRun: opts.dryRun
+      });
+      results.push(result);
+    } catch (err) {
+      logger.warn(
+        'SYSTEM',
+        'Worktree adoption failed for parent repo (continuing)',
+        { repoPath, error: err instanceof Error ? err.message : String(err) }
+      );
+    }
+  }
+
+  return results;
+}
@@ -1,93 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22310 | 9:46 PM | 🟣 | Complete Hook Lifecycle Documentation Generated | ~603 |
-| #22305 | 9:45 PM | 🔵 | Session Summary Storage and Status Lifecycle | ~472 |
-| #22304 | " | 🔵 | Session Creation Idempotency and Observation Storage | ~481 |
-| #22303 | " | 🔵 | SessionStore CRUD Operations for Hook Integration | ~392 |
-| #22300 | 9:44 PM | 🔵 | SessionStore Database Management and Schema Migrations | ~455 |
-| #22299 | " | 🔵 | Database Schema and Entity Types | ~460 |
-| #21976 | 5:24 PM | 🟣 | storeObservation Saves tool_use_id to Database | ~298 |
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23808 | 10:42 PM | 🔵 | migrations.ts Already Migrated to bun:sqlite | ~312 |
-| #23807 | " | 🔵 | SessionSearch.ts Already Migrated to bun:sqlite | ~321 |
-| #23805 | " | 🔵 | Database.ts Already Migrated to bun:sqlite | ~290 |
-| #23784 | 9:59 PM | ✅ | SessionStore.ts db.pragma() Converted to db.query().all() Pattern | ~198 |
-| #23783 | 9:58 PM | ✅ | SessionStore.ts Migration004 Multi-Statement db.exec() Converted to db.run() | ~220 |
-| #23782 | " | ✅ | SessionStore.ts initializeSchema() db.exec() Converted to db.run() | ~197 |
-| #23781 | " | ✅ | SessionStore.ts Constructor PRAGMA Calls Converted to db.run() | ~215 |
-| #23780 | " | ✅ | SessionStore.ts Type Annotation Updated | ~183 |
-| #23779 | " | ✅ | SessionStore.ts Import Updated to bun:sqlite | ~237 |
-| #23778 | 9:57 PM | ✅ | Database.ts Import Updated to bun:sqlite | ~177 |
-| #23777 | " | 🔵 | SessionStore.ts Current Implementation - better-sqlite3 Import and API Usage | ~415 |
-| #23776 | " | 🔵 | migrations.ts Current Implementation - better-sqlite3 Import | ~285 |
-| #23775 | " | 🔵 | Database.ts Current Implementation - better-sqlite3 Import | ~286 |
-| #23774 | " | 🔵 | SessionSearch.ts Current Implementation - better-sqlite3 Import | ~309 |
-| #23671 | 8:36 PM | 🔵 | getUserPromptsByIds Method Implementation with Filtering and Ordering | ~326 |
-| #23670 | " | 🔵 | getUserPromptsByIds Method Location in SessionStore | ~145 |
-| #23635 | 8:10 PM | 🔴 | Fixed SessionStore.ts Concepts Filter SQL Parameter Bug | ~297 |
-| #23634 | " | 🔵 | SessionStore.ts Concepts Filter Bug Confirmed at Line 849 | ~356 |
-| #23522 | 5:27 PM | 🔵 | Complete TypeScript Type Definitions for Database Entities | ~433 |
-| #23521 | " | 🔵 | Database Schema Structure with 7 Migration Versions | ~461 |
-
-### Dec 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #29868 | 8:19 PM | 🔵 | SessionStore Architecture Review for Mode Metadata Addition | ~350 |
-| #29243 | 12:13 AM | 🔵 | Observations Table Schema Migration: Text Field Made Nullable | ~496 |
-| #29241 | 12:12 AM | 🔵 | Migration001: Core Schema for Sessions, Memories, Overviews, Diagnostics, Transcripts | ~555 |
-| #29238 | 12:11 AM | 🔵 | Observation Type Schema Evolution: Five to Six Types | ~331 |
-| #29237 | " | 🔵 | SQLite SessionStore with Schema Migrations and WAL Mode | ~520 |
-
-### Dec 21, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #31622 | 8:26 PM | 🔄 | Completed SessionStore logging standardization | ~270 |
-| #31621 | " | 🔄 | Standardized error logging for boundary timestamps query | ~253 |
-| #31620 | " | 🔄 | Standardized error logging in getTimelineAroundObservation | ~252 |
-| #31619 | " | 🔄 | Replaced console.log with logger.debug in SessionStore | ~263 |
-
-### Dec 27, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #33213 | 9:04 PM | 🔵 | SessionStore Implements KISS Session ID Threading via INSERT OR IGNORE Pattern | ~673 |
-
-### Dec 28, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #33548 | 10:59 PM | ✅ | Reverted memory_session_id NULL Initialization to contentSessionId Placeholder | ~421 |
-| #33546 | 10:57 PM | 🔴 | Fixed createSDKSession to Initialize memory_session_id as NULL | ~406 |
-| #33545 | " | 🔵 | createSDKSession Sets memory_session_id Equal to content_session_id Initially | ~378 |
-| #33544 | " | 🔵 | SessionStore Migration 17 Already Renamed Session ID Columns | ~451 |
-
-### Jan 2, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36028 | 9:20 PM | 🔄 | Try-Catch Block Removed from Database Migration | ~291 |
-
-### Jan 3, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36653 | 11:03 PM | 🔵 | storeObservation Method Signature Shows Parameter Named memorySessionId | ~474 |
-| #36652 | " | 🔵 | createSDKSession Implementation Confirms NULL Initialization With Security Rationale | ~488 |
-| #36650 | 11:02 PM | 🔵 | Phase 1 Analysis Reveals Implementation-Test Mismatch on NULL vs Placeholder Initialization | ~687 |
-| #36649 | " | 🔵 | SessionStore Implementation Reveals NULL-Based Memory Session ID Initialization Pattern | ~770 |
-| #36175 | 6:52 PM | ✅ | MigrationRunner Re-exported from Migrations.ts | ~405 |
-| #36172 | " | 🔵 | Migrations.ts Contains Legacy Migration System | ~650 |
-| #36163 | 6:48 PM | 🔵 | SessionStore Method Inventory and Extraction Boundaries | ~692 |
-| #36162 | 6:47 PM | 🔵 | SessionStore Architecture and Migration History | ~593 |
-</claude-mem-context>
@@ -1,5 +1,5 @@
 import { Database } from 'bun:sqlite';
-import { DATA_DIR, DB_PATH, ensureDir } from '../../shared/paths.js';
+import { DATA_DIR, DB_PATH, ensureDir, OBSERVER_SESSIONS_PROJECT } from '../../shared/paths.js';
 import { logger } from '../../utils/logger.js';
 import {
  TableColumnInfo,
@@ -65,6 +65,7 @@ export class SessionStore {
    this.addSessionCustomTitleColumn();
    this.addSessionPlatformSourceColumn();
    this.addObservationModelColumns();
+    this.ensureMergedIntoProjectColumns();
  }

  /**
@@ -944,6 +945,36 @@ export class SessionStore {
    this.db.prepare('INSERT OR IGNORE INTO schema_versions (version, applied_at) VALUES (?, ?)').run(26, new Date().toISOString());
  }

+  /**
+   * Ensure merged_into_project columns + indices exist on observations and session_summaries.
+   *
+   * Self-idempotent via PRAGMA table_info guard — does NOT bump schema_versions.
+   * Mirrors MigrationRunner.ensureMergedIntoProjectColumns so bundled artifacts
+   * that embed SessionStore (e.g. context-generator.cjs) stay schema-consistent
+   * with the standalone migration path.
+   */
+  private ensureMergedIntoProjectColumns(): void {
+    const obsCols = this.db
+      .query('PRAGMA table_info(observations)')
+      .all() as TableColumnInfo[];
+    if (!obsCols.some(c => c.name === 'merged_into_project')) {
+      this.db.run('ALTER TABLE observations ADD COLUMN merged_into_project TEXT');
+    }
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_observations_merged_into ON observations(merged_into_project)'
+    );
+
+    const sumCols = this.db
+      .query('PRAGMA table_info(session_summaries)')
+      .all() as TableColumnInfo[];
+    if (!sumCols.some(c => c.name === 'merged_into_project')) {
+      this.db.run('ALTER TABLE session_summaries ADD COLUMN merged_into_project TEXT');
+    }
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_summaries_merged_into ON session_summaries(merged_into_project)'
+    );
+  }
+
  /**
   * Update the memory session ID for a session
   * Called by SDKAgent when it captures the session ID from the first SDK message
@@ -1192,8 +1223,9 @@ export class SessionStore {
      SELECT DISTINCT project
      FROM sdk_sessions
      WHERE project IS NOT NULL AND project != ''
+        AND project != ?
    `;
-    const params: unknown[] = [];
+    const params: unknown[] = [OBSERVER_SESSIONS_PROJECT];

    if (normalizedPlatformSource) {
      query += ' AND COALESCE(platform_source, ?) = ?';
@@ -1218,9 +1250,10 @@ export class SessionStore {
        MAX(started_at_epoch) as latest_epoch
      FROM sdk_sessions
      WHERE project IS NOT NULL AND project != ''
+        AND project != ?
      GROUP BY COALESCE(platform_source, '${DEFAULT_PLATFORM_SOURCE}'), project
      ORDER BY latest_epoch DESC
-    `).all() as Array<{ platform_source: string; project: string; latest_epoch: number }>;
+    `).all(OBSERVER_SESSIONS_PROJECT) as Array<{ platform_source: string; project: string; latest_epoch: number }>;

    const projects: string[] = [];
    const seenProjects = new Set<string>();
@@ -37,6 +37,7 @@ export class MigrationRunner {
    this.addSessionCustomTitleColumn();
    this.createObservationFeedbackTable();
    this.addSessionPlatformSourceColumn();
+    this.ensureMergedIntoProjectColumns();
  }

  /**
@@ -922,4 +923,33 @@ export class MigrationRunner {

    this.db.prepare('INSERT OR IGNORE INTO schema_versions (version, applied_at) VALUES (?, ?)').run(25, new Date().toISOString());
  }
+
+  /**
+   * Ensure merged_into_project columns + indices exist on observations and session_summaries.
+   *
+   * Self-idempotent via PRAGMA table_info guard — does NOT bump schema_versions.
+   * Supports merged-worktree adoption: a nullable pointer that lets a worktree's rows
+   * be surfaced under the parent project's observation list without data movement.
+   */
+  private ensureMergedIntoProjectColumns(): void {
+    const obsCols = this.db
+      .query('PRAGMA table_info(observations)')
+      .all() as TableColumnInfo[];
+    if (!obsCols.some(c => c.name === 'merged_into_project')) {
+      this.db.run('ALTER TABLE observations ADD COLUMN merged_into_project TEXT');
+    }
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_observations_merged_into ON observations(merged_into_project)'
+    );
+
+    const sumCols = this.db
+      .query('PRAGMA table_info(session_summaries)')
+      .all() as TableColumnInfo[];
+    if (!sumCols.some(c => c.name === 'merged_into_project')) {
+      this.db.run('ALTER TABLE session_summaries ADD COLUMN merged_into_project TEXT');
+    }
+    this.db.run(
+      'CREATE INDEX IF NOT EXISTS idx_summaries_merged_into ON session_summaries(merged_into_project)'
+    );
+  }
 }
@@ -6,7 +6,7 @@
 import { createHash } from 'crypto';
 import { Database } from 'bun:sqlite';
 import { logger } from '../../../utils/logger.js';
-import { getCurrentProjectName } from '../../../shared/paths.js';
+import { getProjectContext } from '../../../utils/project-name.js';
 import type { ObservationInput, StoreObservationResult } from './types.js';

 /** Deduplication window: observations with the same content hash within this window are skipped */
@@ -62,7 +62,7 @@ export function storeObservation(
  const timestampIso = new Date(timestampEpoch).toISOString();

  // Guard against empty project string (race condition where project isn't set yet)
-  const resolvedProject = project || getCurrentProjectName();
+  const resolvedProject = project || getProjectContext(process.cwd()).primary;

  // Content-hash deduplication
  const contentHash = computeObservationContentHash(memorySessionId, observation.title, observation.narrative);
@@ -8,6 +8,7 @@
 import type { Database } from 'bun:sqlite';
 import type { ObservationRecord, SessionSummaryRecord, UserPromptRecord } from '../../../types/database.js';
 import { logger } from '../../../utils/logger.js';
+import { OBSERVER_SESSIONS_PROJECT } from '../../../shared/paths.js';

 /**
 * Timeline result containing observations, sessions, and prompts within a time window
@@ -210,9 +211,10 @@ export function getAllProjects(db: Database): string[] {
    SELECT DISTINCT project
    FROM sdk_sessions
    WHERE project IS NOT NULL AND project != ''
+      AND project != ?
    ORDER BY project ASC
  `);

-  const rows = stmt.all() as Array<{ project: string }>;
+  const rows = stmt.all(OBSERVER_SESSIONS_PROJECT) as Array<{ project: string }>;
  return rows.map(row => row.project);
 }
@@ -28,6 +28,7 @@ interface StoredObservation {
  id: number;
  memory_session_id: string;
  project: string;
+  merged_into_project: string | null;
  text: string | null;
  type: string;
  title: string | null;
@@ -47,6 +48,7 @@ interface StoredSummary {
  id: number;
  memory_session_id: string;
  project: string;
+  merged_into_project: string | null;
  request: string | null;
  investigated: string | null;
  learned: string | null;
@@ -129,11 +131,12 @@ export class ChromaSync {
    const files_read = parseFileList(obs.files_read);
    const files_modified = parseFileList(obs.files_modified);

-    const baseMetadata: Record<string, string | number> = {
+    const baseMetadata: Record<string, string | number | null> = {
      sqlite_id: obs.id,
      doc_type: 'observation',
      memory_session_id: obs.memory_session_id,
      project: obs.project,
+      merged_into_project: obs.merged_into_project ?? null,
      created_at_epoch: obs.created_at_epoch,
      type: obs.type || 'discovery',
      title: obs.title || 'Untitled'
@@ -190,11 +193,12 @@ export class ChromaSync {
  private formatSummaryDocs(summary: StoredSummary): ChromaDocument[] {
    const documents: ChromaDocument[] = [];

-    const baseMetadata: Record<string, string | number> = {
+    const baseMetadata: Record<string, string | number | null> = {
      sqlite_id: summary.id,
      doc_type: 'session_summary',
      memory_session_id: summary.memory_session_id,
      project: summary.project,
+      merged_into_project: summary.merged_into_project ?? null,
      created_at_epoch: summary.created_at_epoch,
      prompt_number: summary.prompt_number || 0
    };
@@ -346,6 +350,7 @@ export class ChromaSync {
      id: observationId,
      memory_session_id: memorySessionId,
      project: project,
+      merged_into_project: null,
      text: null, // Legacy field, not used
      type: obs.type,
      title: obs.title,
@@ -390,6 +395,7 @@ export class ChromaSync {
      id: summaryId,
      memory_session_id: memorySessionId,
      project: project,
+      merged_into_project: null,
      request: summary.request,
      investigated: summary.investigated,
      learned: summary.learned,
@@ -830,6 +836,72 @@ export class ChromaSync {
    }
  }

+  /**
+   * Stamp `merged_into_project` on every Chroma document whose metadata
+   * `sqlite_id` is in the provided set. Used by the worktree adoption engine
+   * to keep Chroma's metadata in lockstep with SQLite after a parent branch
+   * absorbs a worktree branch via merge.
+   *
+   * Batched: fetches docs by `sqlite_id IN sqliteIds`, rewrites metadata with
+   * the new field, and calls `chroma_update_documents` once per page of up to
+   * BATCH_SIZE ids. Idempotent — re-running with the same value is a no-op
+   * because the write doesn't depend on the prior value.
+   */
+  async updateMergedIntoProject(
+    sqliteIds: number[],
+    mergedIntoProject: string
+  ): Promise<void> {
+    if (sqliteIds.length === 0) return;
+
+    await this.ensureCollectionExists();
+    const chromaMcp = ChromaMcpManager.getInstance();
+
+    let totalPatched = 0;
+
+    // Chunk the sqlite_id set to keep each Chroma call bounded.
+    for (let i = 0; i < sqliteIds.length; i += this.BATCH_SIZE) {
+      const idBatch = sqliteIds.slice(i, i + this.BATCH_SIZE);
+
+      const existing = await chromaMcp.callTool('chroma_get_documents', {
+        collection_name: this.collectionName,
+        where: { sqlite_id: { $in: idBatch } },
+        include: ['metadatas']
+      }) as { ids?: string[]; metadatas?: Array<Record<string, any> | null> };
+
+      const docIds: string[] = existing?.ids ?? [];
+      if (docIds.length === 0) continue;
+
+      const metadatas = (existing?.metadatas ?? []).map(m => {
+        // Merge old metadata with the new field, then filter out null/undefined/''
+        // to match the sanitization other callTool sites apply (chroma-mcp
+        // rejects null values in metadata).
+        const merged: Record<string, any> = {
+          ...(m ?? {}),
+          merged_into_project: mergedIntoProject
+        };
+        return Object.fromEntries(
+          Object.entries(merged).filter(
+            ([, v]) => v !== null && v !== undefined && v !== ''
+          )
+        );
+      });
+
+      await chromaMcp.callTool('chroma_update_documents', {
+        collection_name: this.collectionName,
+        ids: docIds,
+        metadatas
+      });
+      totalPatched += docIds.length;
+    }
+
+    logger.info('CHROMA_SYNC', 'merged_into_project metadata patched', {
+      collection: this.collectionName,
+      mergedIntoProject,
+      sqliteIdCount: sqliteIds.length,
+      chromaDocsPatched: totalPatched
+    });
+  }
+
  /**
   * Close the ChromaSync instance
   * ChromaMcpManager is a singleton and manages its own lifecycle
@@ -45,6 +45,7 @@ import {
  getPlatformTimeout,
  aggressiveStartupCleanup,
  runOneTimeChromaMigration,
+  runOneTimeCwdRemap,
  cleanStalePidFile,
  isProcessAlive,
  spawnDaemon,
@@ -58,6 +59,7 @@ import {
  httpShutdown
 } from './infrastructure/HealthMonitor.js';
 import { performGracefulShutdown } from './infrastructure/GracefulShutdown.js';
+import { adoptMergedWorktrees, adoptMergedWorktreesForAllKnownRepos } from './infrastructure/WorktreeAdoption.js';

 // Server imports
 import { Server } from './server/Server.js';
@@ -359,6 +361,34 @@ export class WorkerService {
        runOneTimeChromaMigration();
      }

+      // One-time remap of pre-worktree project names using pending_messages.cwd.
+      // Must run before dbManager.initialize() so we don't hold the DB open.
+      runOneTimeCwdRemap();
+
+      // Stamp merged worktrees so their observations surface under the parent
+      // project. Runs every startup (not marker-gated) because git state evolves
+      // and the engine is fully idempotent. Must also precede dbManager.initialize().
+      //
+      // The worker daemon is spawned with cwd=marketplace-plugin-dir (not a git
+      // repo), so we can't seed adoption with process.cwd(). Instead, discover
+      // parent repos from recorded pending_messages.cwd values.
+      try {
+        const adoptions = await adoptMergedWorktreesForAllKnownRepos({});
+        for (const adoption of adoptions) {
+          if (adoption.adoptedObservations > 0 || adoption.adoptedSummaries > 0 || adoption.chromaUpdates > 0) {
+            logger.info('SYSTEM', 'Merged worktrees adopted on startup', adoption);
+          }
+          if (adoption.errors.length > 0) {
+            logger.warn('SYSTEM', 'Worktree adoption had per-branch errors', {
+              repoPath: adoption.repoPath,
+              errors: adoption.errors
+            });
+          }
+        }
+      } catch (err) {
+        logger.error('SYSTEM', 'Worktree adoption failed (non-fatal)', {}, err as Error);
+      }
+
      // Initialize ChromaMcpManager only if Chroma is enabled
      const chromaEnabled = settings.CLAUDE_MEM_CHROMA_ENABLED !== 'false';
      if (chromaEnabled) {
@@ -1187,6 +1217,45 @@ async function main() {
      break;
    }

+    case 'adopt': {
+      const dryRun = process.argv.includes('--dry-run');
+      const branchIndex = process.argv.indexOf('--branch');
+      const branchValue = branchIndex !== -1 ? process.argv[branchIndex + 1] : undefined;
+      if (branchIndex !== -1 && (!branchValue || branchValue.startsWith('--'))) {
+        console.error('Usage: adopt [--dry-run] [--branch <branch>] [--cwd <path>]');
+        process.exit(1);
+      }
+      const onlyBranch = branchValue;
+      // Honor an explicit --cwd override so the NPX CLI can pass through the
+      // user's working directory (the spawn sets cwd to the marketplace dir).
+      const cwdIndex = process.argv.indexOf('--cwd');
+      const cwdValue = cwdIndex !== -1 ? process.argv[cwdIndex + 1] : undefined;
+      if (cwdIndex !== -1 && (!cwdValue || cwdValue.startsWith('--'))) {
+        console.error('Usage: adopt [--dry-run] [--branch <branch>] [--cwd <path>]');
+        process.exit(1);
+      }
+      const repoPath = cwdValue ?? process.cwd();
+
+      const result = await adoptMergedWorktrees({ repoPath, dryRun, onlyBranch });
+
+      const tag = result.dryRun ? '(dry-run)' : '(applied)';
+      console.log(`\nWorktree adoption ${tag}`);
+      console.log(`  Parent project:       ${result.parentProject || '(unknown)'}`);
+      console.log(`  Repo:                 ${result.repoPath}`);
+      console.log(`  Worktrees scanned:    ${result.scannedWorktrees}`);
+      console.log(`  Merged branches:      ${result.mergedBranches.join(', ') || '(none)'}`);
+      console.log(`  Observations adopted: ${result.adoptedObservations}`);
+      console.log(`  Summaries adopted:    ${result.adoptedSummaries}`);
+      console.log(`  Chroma docs updated:  ${result.chromaUpdates}`);
+      if (result.chromaFailed > 0) {
+        console.log(`  Chroma sync failures: ${result.chromaFailed} (will retry on next run)`);
+      }
+      for (const err of result.errors) {
+        console.log(`  ! ${err.worktree}: ${err.error}`);
+      }
+      process.exit(0);
+    }
+
    case '--daemon':
    default: {
      // GUARD 1: Refuse to start if another worker is already alive (PID check).
@@ -124,6 +124,7 @@ export interface Observation {
  id: number;
  memory_session_id: string;  // Renamed from sdk_session_id
  project: string;
+  merged_into_project: string | null;
  platform_source: string;
  type: string;
  title: string;
@@ -1,123 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23673 | 8:36 PM | ✅ | Add Project Filter Parameter to Session and Prompt Hydration in Search | ~306 |
-| #23596 | 5:54 PM | ⚖️ | Import/Export Bug Fix Priority and Scope | ~415 |
-| #23595 | 5:53 PM | 🔴 | SearchManager Returns Wrong Format for Empty Results | ~320 |
-| #23594 | " | 🔵 | SearchManager Search Method Control Flow | ~313 |
-| #23591 | 5:51 PM | 🔵 | SearchManager JSON Response Structure | ~231 |
-| #23590 | " | 🔵 | Import/Export Feature Status Review | ~490 |
-| #23583 | 5:50 PM | 🔵 | SearchManager Hybrid Search Architecture | ~495 |
-
-### Dec 13, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #25191 | 8:04 PM | 🔵 | ChromaSync Instantiated in DatabaseManager Constructor | ~315 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26263 | 8:32 PM | 🔵 | SearchManager Timeline Methods Use Rich Formatting, Search Method Uses Flat Tables | ~464 |
-| #26243 | 8:29 PM | 🔵 | FormattingService Provides Basic Table Format Without Dates or File Grouping | ~390 |
-| #26240 | " | 🔵 | SearchManager Formats Results as Tables, Timeline Uses Rich Date-Grouped Format | ~416 |
-| #26108 | 7:43 PM | ✅ | changes() Method Format Logic Removed | ~401 |
-| #26107 | " | ✅ | changes() Method Format Parameter Removed | ~317 |
-| #26106 | 7:42 PM | ✅ | decisions() Method Format Logic Removed | ~405 |
-| #26105 | " | ✅ | decisions() Method Format Parameter Removed | ~310 |
-| #26104 | " | ✅ | Main search() Method Format Handling Removed | ~430 |
-| #26103 | 7:41 PM | ✅ | FormattingService.ts Rewritten to Table Format | ~457 |
-| #26102 | " | 🔵 | SearchManager.ts Format Parameter Removal Status | ~478 |
-
-### Dec 15, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #27043 | 6:04 PM | 🔵 | Subagent confirms no version switcher UI exists, only orphaned backend infrastructure | ~539 |
-| #27041 | 6:03 PM | 🔵 | Branch switching code isolated to two backend files, no frontend UI components | ~473 |
-| #27037 | 6:02 PM | 🔵 | Branch switching functionality exists in SettingsRoutes with UI switcher removal intent | ~463 |
-
-### Dec 16, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #27727 | 5:45 PM | 🔵 | SearchManager returns raw data arrays when format=json is specified | ~349 |
-
-### Dec 17, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #28473 | 4:25 PM | 🔵 | PaginationHelper LIMIT+1 Trick and Project Path Sanitization | ~499 |
-| #28458 | 4:24 PM | 🔵 | SDK Agent Observer-Only Event-Driven Query Loop | ~513 |
-| #28455 | " | 🔵 | Event-Driven Session Manager with Zero-Latency Queuing | ~566 |
-
-### Dec 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #29240 | 12:12 AM | 🔵 | SDK Agent Event-Driven Query Loop with Tool Restrictions | ~507 |
-
-### Dec 20, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #31100 | 8:01 PM | 🔵 | Summary and Memory Message Generation in SDK Agent | ~324 |
-
-### Dec 25, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32616 | 8:43 PM | 🔵 | Comprehensive analysis of "enable billing" setting and its impact on rate limiting | ~533 |
-| #32599 | 8:40 PM | 🔄 | Added validation and explicit default for Gemini model configuration | ~393 |
-| #32598 | " | 🔵 | Gemini configuration loaded from settings or environment variables | ~363 |
-| #32591 | 8:38 PM | 🔴 | Removed Unsupported Gemini Model from Agent | ~282 |
-| #32583 | " | 🔵 | Gemini Agent Implementation Details | ~434 |
-| #32543 | 7:29 PM | 🔄 | Rate limiting applied conditionally based on billing status | ~164 |
-| #32542 | " | 🔄 | Query Gemini now accepts billing status | ~163 |
-| #32541 | " | 🔄 | Gemini config now includes billing status | ~182 |
-| #32540 | " | 🔄 | Rate limiting logic refactored for Gemini billing | ~164 |
-
-### Dec 26, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32949 | 10:55 PM | 🔵 | Complete settings persistence flow for Xiaomi MIMO v2 Flash model | ~320 |
-| #32948 | 10:53 PM | 🔵 | OpenRouterAgent uses CLAUDE_MEM_OPENROUTER_MODEL setting with Xiaomi as default | ~183 |
-
-### Dec 27, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #33215 | 9:06 PM | 🔵 | SessionManager Implements Event-Driven Lifecycle with Database-First Persistence and Auto-Initialization | ~853 |
-| #33214 | " | 🔵 | SDKAgent Implements Event-Driven Query Loop with Init/Continuation Prompt Selection | ~769 |
-
-### Dec 28, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #33551 | 11:00 PM | 🔵 | GeminiAgent Does Not Implement Resume Functionality | ~307 |
-| #33550 | " | 🔵 | OpenRouterAgent Does Not Implement Resume Functionality | ~294 |
-| #33549 | 10:59 PM | 🔴 | SDKAgent Now Checks memorySessionId Differs From contentSessionId Before Resume | ~419 |
-| #33547 | " | 🔵 | All Agents Call storeObservation with contentSessionId Instead of memorySessionId | ~407 |
-| #33543 | 10:56 PM | 🔵 | SDKAgent Already Implements Memory Session ID Capture and Resume Logic | ~467 |
-| #33542 | " | 🔵 | SessionManager Already Uses Renamed Session ID Fields | ~390 |
-
-### Dec 30, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #34504 | 2:31 PM | 🔵 | SDKAgent V2 Message Handling and Processing Flow Detailed | ~583 |
-| #34459 | 2:23 PM | 🔵 | Complete SDKAgent V2 Architecture with Comprehensive Message Processing | ~619 |
-| #34453 | 2:21 PM | 🔵 | Memory Agent Configured as Observer-Only | ~379 |
-
-### Jan 4, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36853 | 1:49 AM | 🔵 | GeminiAgent Implementation Reviewed for Model Support | ~555 |
-</claude-mem-context>
@@ -9,6 +9,7 @@

 import { DatabaseManager } from './DatabaseManager.js';
 import { logger } from '../../utils/logger.js';
+import { OBSERVER_SESSIONS_PROJECT } from '../../shared/paths.js';
 import type { PaginatedResult, Observation, Summary, UserPrompt } from '../worker-types.js';

 export class PaginationHelper {
@@ -24,15 +25,17 @@ export class PaginationHelper {
   * Uses first occurrence of project name from left (project root)
   */
  private stripProjectPath(filePath: string, projectName: string): string {
-    const marker = `/${projectName}/`;
+    // Composite names ("parent/worktree") don't appear in on-disk paths for
+    // standard git worktrees — only the checkout basename does. Match on the
+    // leaf segment so the heuristic works regardless of worktree layout.
+    const leaf = projectName.includes('/') ? projectName.split('/').pop()! : projectName;
+    const marker = `/${leaf}/`;
    const index = filePath.indexOf(marker);

    if (index !== -1) {
-      // Strip everything before and including the project name
      return filePath.substring(index + marker.length);
    }

-    // Fallback: return original path if project name not found
    return filePath;
  }

@@ -78,6 +81,7 @@ export class PaginationHelper {
        o.id,
        o.memory_session_id,
        o.project,
+        o.merged_into_project,
        COALESCE(s.platform_source, 'claude') as platform_source,
        o.type,
        o.title,
@@ -98,8 +102,14 @@ export class PaginationHelper {
    const conditions: string[] = [];

    if (project) {
-      conditions.push('o.project = ?');
-      params.push(project);
+      // Include adopted merged-worktree rows so the parent project's view
+      // surfaces observations that originated under its merged children.
+      conditions.push('(o.project = ? OR o.merged_into_project = ?)');
+      params.push(project, project);
+    } else {
+      // Hide internal observer-session rows from the unfiltered UI list.
+      conditions.push('o.project != ?');
+      params.push(OBSERVER_SESSIONS_PROJECT);
    }
    if (platformSource) {
      conditions.push(`COALESCE(s.platform_source, 'claude') = ?`);
@@ -154,8 +164,13 @@ export class PaginationHelper {
    const conditions: string[] = [];

    if (project) {
-      conditions.push('ss.project = ?');
-      params.push(project);
+      // Include adopted merged-worktree summaries so the parent project's view
+      // surfaces rows that originated under its merged children.
+      conditions.push('(ss.project = ? OR ss.merged_into_project = ?)');
+      params.push(project, project);
+    } else {
+      // Hide internal observer-session rows from the unfiltered UI list.
+      conditions.push("ss.project != 'observer-sessions'");
    }

    if (platformSource) {
@@ -207,6 +222,9 @@ export class PaginationHelper {
    if (project) {
      conditions.push('s.project = ?');
      params.push(project);
+    } else {
+      // Hide internal observer-session rows from the unfiltered UI list.
+      conditions.push("s.project != 'observer-sessions'");
    }

    if (platformSource) {
@@ -412,11 +412,22 @@ export function createPidCapturingSpawn(sessionDbId: number) {
    const useCmdWrapper = process.platform === 'win32' && spawnOptions.command.endsWith('.cmd');
    const env = sanitizeEnv(spawnOptions.env ?? process.env);

-    // Filter empty string args: Bun's spawn() silently drops empty strings from argv,
-    // causing subsequent flags to be consumed as values for the preceding flag.
-    // The Agent SDK may produce empty-string args (e.g., settingSources defaults to []
-    // which joins to ""). Node preserves these, but Bun drops them, breaking CLI parsing.
-    const args = spawnOptions.args.filter(arg => arg !== '');
+    // Filter empty string args AND their preceding flag (Issue #2049).
+    // The Agent SDK emits ["--setting-sources", ""] when settingSources defaults to [].
+    // Simply dropping "" leaves an orphan --setting-sources that consumes the next
+    // flag (e.g. --permission-mode) as its value, crashing Claude Code 2.1.109+ with
+    // "Invalid setting source: --permission-mode". Drop the flag too so the SDK
+    // default (no setting sources) is preserved by omission.
+    const args: string[] = [];
+    for (const arg of spawnOptions.args) {
+      if (arg === '') {
+        if (args.length > 0 && args[args.length - 1].startsWith('--')) {
+          args.pop();
+        }
+        continue;
+      }
+      args.push(arg);
+    }

    const child = useCmdWrapper
      ? spawn('cmd.exe', ['/d', '/c', spawnOptions.command, ...args], {
@@ -13,7 +13,6 @@
 * - TimelineBuilder: Timeline construction
 */

-import { basename } from 'path';
 import { SessionSearch } from '../sqlite/SessionSearch.js';
 import { SessionStore } from '../sqlite/SessionStore.js';
 import { ChromaSync } from '../sync/ChromaSync.js';
@@ -22,6 +21,7 @@ import { TimelineService } from './TimelineService.js';
 import type { TimelineItem } from './TimelineService.js';
 import type { ObservationSearchResult, SessionSummarySearchResult, UserPromptSearchResult } from '../sqlite/types.js';
 import { logger } from '../../utils/logger.js';
+import { getProjectContext } from '../../utils/project-name.js';
 import { formatDate, formatTime, formatDateTime, extractFirstFile, groupByDate, estimateTokens } from '../../shared/timeline-formatting.js';
 import { ModeManager } from '../domain/ModeManager.js';

@@ -170,8 +170,16 @@ export class SearchManager {
      // Include project in the Chroma where clause to scope vector search.
      // Without this, larger projects dominate the top-N results and smaller
      // projects get crowded out before the post-hoc SQLite filter.
+      // Match both native-provenance rows (project) and adopted merged-worktree
+      // rows (merged_into_project) so a parent-project query surfaces its
+      // merged children's observations too.
      if (options.project) {
-        const projectFilter = { project: options.project };
+        const projectFilter = {
+          $or: [
+            { project: options.project },
+            { merged_into_project: options.project }
+          ]
+        };
        whereFilter = whereFilter
          ? { $and: [whereFilter, projectFilter] }
          : projectFilter;
@@ -1319,7 +1327,7 @@ export class SearchManager {
   * Tool handler: get_recent_context
   */
  async getRecentContext(args: any): Promise<any> {
-    const project = args.project || basename(process.cwd());
+    const project = args.project || getProjectContext(process.cwd()).primary;
    const limit = args.limit || 3;

    const sessions = this.sessionStore.getRecentSessionsWithStatus(project, limit);
@@ -85,27 +85,6 @@ export async function processAgentResponse(
  // Convert nullable fields to empty strings for storeSummary (if summary exists)
  const summaryForStore = normalizeSummaryForStorage(summary);

-  // Fallback: When summary parse fails but observations exist, salvage a synthetic summary.
-  // Fixes Issue #1312: AI sometimes returns <observation> instead of <summary> despite clear instructions.
-  // Observations are stored normally; this only affects the session summary.
-  let finalSummaryForStore = summaryForStore;
-  if (!summaryForStore && observations.length > 0) {
-    const primary = observations[0];
-    finalSummaryForStore = {
-      request: primary.title || `Session observations (${observations.length} items)`,
-      investigated: primary.narrative || primary.facts?.join('; ') || '',
-      learned: primary.facts?.join('; ') || '',
-      completed: primary.type === 'feature' || primary.type === 'bugfix' ? (primary.title || '') : '',
-      next_steps: '',
-      notes: `[Salvaged from ${observations.length} observation(s)] AI returned <observation> instead of <summary>`
-    };
-    logger.warn('PARSER', `SALVAGED summary from ${observations.length} observation(s) — AI did not output <summary> tags`, {
-      sessionId: session.sessionDbId,
-      agentName,
-      observationIds: observations.map(o => o.title).filter(Boolean).slice(0, 3)
-    });
-  }
-
  // Get session store for atomic transaction
  const sessionStore = dbManager.getSessionStore();

@@ -123,7 +102,7 @@ export async function processAgentResponse(
  sessionStore.ensureMemorySessionIdRegistered(session.sessionDbId, session.memorySessionId);

  // Log pre-storage with session ID chain for verification
-  logger.info('DB', `STORING | sessionDbId=${session.sessionDbId} | memorySessionId=${session.memorySessionId} | obsCount=${observations.length} | hasSummary=${!!finalSummaryForStore}`, {
+  logger.info('DB', `STORING | sessionDbId=${session.sessionDbId} | memorySessionId=${session.memorySessionId} | obsCount=${observations.length} | hasSummary=${!!summaryForStore}`, {
    sessionId: session.sessionDbId,
    memorySessionId: session.memorySessionId
  });
@@ -134,7 +113,7 @@ export async function processAgentResponse(
    session.memorySessionId,
    session.project,
    observations,
-    finalSummaryForStore,
+    summaryForStore,
    session.lastPromptNumber,
    discoveryTokens,
    originalTimestamp ?? undefined,
@@ -178,7 +157,7 @@ export async function processAgentResponse(
  // Sync and broadcast summary if present
  await syncAndBroadcastSummary(
    summary,
-    finalSummaryForStore,
+    summaryForStore,
    result,
    session,
    dbManager,
@@ -1,113 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Nov 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6295 | 1:18 PM | 🔵 | Path Configuration Structure for claude-mem | ~305 |
-
-### Dec 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20730 | 9:06 PM | 🔵 | Path Configuration Module with ESM/CJS Compatibility | ~578 |
-| #20718 | 9:00 PM | 🔵 | Worker Service Auto-Start and Health Check System | ~448 |
-| #20410 | 7:21 PM | 🔵 | Path utilities provide cross-runtime directory management with Claude integration support | ~478 |
-| #20409 | 7:20 PM | 🔵 | Worker utilities provide automatic PM2 startup with health checking and port configuration | ~479 |
-
-### Dec 9, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23141 | 6:42 PM | 🔵 | Located getSettingsPath Function in paths.ts | ~261 |
-| #23134 | 6:41 PM | ✅ | Set CLAUDE_MEM_SKIP_TOOLS Default Value in SettingsDefaultsManager | ~261 |
-| #23133 | " | ✅ | Added CLAUDE_MEM_SKIP_TOOLS to SettingsDefaults Interface | ~231 |
-| #23131 | 6:40 PM | 🔵 | SettingsDefaultsManager Structure and Configuration Schema | ~363 |
-| #22858 | 2:28 PM | 🔄 | Removed Brittle save.md Validation from paths.ts | ~305 |
-| #22852 | 2:26 PM | 🔵 | Located save.md Validation Logic in paths.ts | ~255 |
-| #22805 | 2:01 PM | 🔵 | Early Settings Silent Failure Point Identified | ~363 |
-| #22803 | " | 🔵 | Worker Utilities Current Implementation Review | ~390 |
-| #22518 | 12:59 AM | 🔵 | Worker Utils StartWorker Implementation Uses Plugin Root for PM2 | ~311 |
-
-### Dec 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #23831 | 11:15 PM | 🔵 | Current hook-error-handler.ts References PM2 | ~277 |
-| #23830 | " | 🔵 | Current worker-utils.ts Implementation Uses PM2 | ~431 |
-| #23812 | 10:49 PM | 🔵 | Current Worker Startup Uses PM2 and PowerShell; Phase 2 Will Replace | ~428 |
-| #23811 | " | 🔵 | Existing Paths Configuration for Phase 2 Reference | ~297 |
-
-### Dec 12, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #24405 | 8:12 PM | 🔵 | PM2 Legacy Cleanup Migration in Worker Startup | ~303 |
-| #24400 | 8:10 PM | 🔵 | Retrieved PM2 Cleanup Implementation Details from Memory | ~355 |
-| #24362 | 7:00 PM | 🟣 | Implemented PM2 Cleanup One-Time Marker in worker-utils.ts | ~376 |
-| #24361 | " | ✅ | Added File System Imports to worker-utils.ts for PM2 Marker | ~263 |
-| #24360 | " | 🔵 | worker-utils.ts Contains PM2 Cleanup Logic Without One-Time Marker | ~390 |
-
-### Dec 13, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #25088 | 7:18 PM | 🟣 | Added CLAUDE_MEM_EMBEDDING_FUNCTION to Settings Interface | ~269 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #26790 | 11:38 PM | 🔴 | Fixed Undefined Port Variable in Error Logger | ~340 |
-| #26789 | " | 🔴 | Fixed Undefined Port Variable in Error Logging | ~316 |
-| #26788 | " | 🔵 | Worker Utils Already Imports Required Dependencies for Implementation | ~283 |
-| #26787 | " | 🟣 | Phase 2 Complete: Pre-Restart Delay Added to Version Mismatch Handler | ~436 |
-| #26786 | " | 🟣 | Phase 2 Complete: Pre-Restart Delay Added to ensureWorkerVersionMatches Function | ~420 |
-| #26785 | 11:37 PM | 🟣 | Phase 1 Complete: PRE_RESTART_SETTLE_DELAY Constant Added to Hook Timeouts | ~351 |
-| #26784 | " | 🟣 | Phase 1 Complete: PRE_RESTART_SETTLE_DELAY Constant Added to HOOK_TIMEOUTS | ~370 |
-| #26783 | " | 🔵 | Hook Constants File Defines Timeout Values and Platform Multiplier | ~452 |
-| #26782 | " | 🔵 | hook-constants.ts Defines Timeout Constants With Windows Platform Multiplier | ~418 |
-| #26766 | 11:30 PM | ⚖️ | Root Cause Identified: Missing Post-Install Worker Restart Trigger in Plugin Update Flow | ~604 |
-| #26765 | " | 🔵 | Explore Agent Confirms Root Cause: No Proactive Worker Restart After Plugin Updates | ~613 |
-| #26732 | 11:25 PM | 🔵 | Worker Utils Implements Version Mismatch Detection and Auto-Restart | ~516 |
-| #26731 | 11:24 PM | 🔵 | ensureWorkerRunning Implementation Shows 2.5 Second Startup Wait With Version Check | ~522 |
-| #25695 | 4:27 PM | 🟣 | Added comprehensive error logging to transcript parser for debugging message extraction failures | ~473 |
-| #25693 | 4:24 PM | 🔵 | Transcript parser extracts messages from JSONL file by scanning backwards for role-specific entries | ~491 |
-
-### Dec 17, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #28464 | 4:25 PM | 🔵 | Platform-Adjusted Hook Timeout Configuration | ~468 |
-| #28461 | " | 🔵 | Dual ESM/CJS Path Resolution System | ~479 |
-| #28452 | 4:23 PM | 🔵 | Worker Version Matching and Auto-Restart System | ~510 |
-
-### Dec 18, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #29797 | 7:09 PM | 🔵 | Settings System Uses CLAUDE_MEM_MODE for Mode Selection | ~353 |
-| #29234 | 12:10 AM | 🔵 | Centralized Settings Management with Environment Defaults | ~394 |
-
-### Dec 20, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #31086 | 7:59 PM | 🔵 | Transcript Parser Extracts Messages from JSONL Hook Files | ~327 |
-| #30939 | 6:57 PM | 🔵 | Worker Utils File Examined for Error Handling Inconsistency | ~393 |
-| #30855 | 6:22 PM | 🔵 | Transcript Parser Content Format Handling Examined | ~406 |
-
-### Dec 25, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32616 | 8:43 PM | 🔵 | Comprehensive analysis of "enable billing" setting and its impact on rate limiting | ~533 |
-| #32538 | 7:28 PM | ✅ | Set default Gemini billing to disabled | ~164 |
-
-### Jan 7, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #38175 | 7:26 PM | 🔵 | Complete Claude-Mem Hook Output Architecture Documented | ~530 |
-</claude-mem-context>
@@ -75,6 +75,10 @@ export const VECTOR_DB_DIR = join(DATA_DIR, 'vector-db');
 // Sessions here won't appear in user's `claude --resume` for their actual projects
 export const OBSERVER_SESSIONS_DIR = join(DATA_DIR, 'observer-sessions');

+// Project name assigned to observer sessions (basename of OBSERVER_SESSIONS_DIR).
+// UI queries filter this out so internal worker sessions don't pollute project lists.
+export const OBSERVER_SESSIONS_PROJECT = basename(OBSERVER_SESSIONS_DIR);
+
 // Claude integration paths
 export const CLAUDE_SETTINGS_PATH = join(CLAUDE_CONFIG_DIR, 'settings.json');
 export const CLAUDE_COMMANDS_DIR = join(CLAUDE_CONFIG_DIR, 'commands');
@@ -1130,6 +1130,19 @@
      color: var(--color-text-muted);
    }

+    /* Merged-into-parent provenance badge */
+    .card-merged-badge {
+      padding: 1px 6px;
+      border-radius: 3px;
+      font-size: 9px;
+      font-weight: 500;
+      letter-spacing: 0.02em;
+      color: var(--color-text-muted);
+      background: var(--color-type-badge-bg);
+      border: 1px solid var(--color-border-primary);
+      opacity: 0.85;
+    }
+
    .summary-card {
      border-color: var(--color-border-summary);
      background: var(--color-bg-summary);
@@ -56,6 +56,11 @@ export function ObservationCard({ observation }: ObservationCardProps) {
            {observation.platform_source || 'claude'}
          </span>
          <span className="card-project">{observation.project}</span>
+          {observation.merged_into_project && (
+            <span className="card-merged-badge" title={`Merged into ${observation.merged_into_project}`}>
+              merged → {observation.merged_into_project}
+            </span>
+          )}
        </div>
        <div className="view-mode-toggles">
          {hasFactsContent && (
@@ -1,9 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Dec 26, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #32982 | 11:04 PM | 🔵 | Read default settings configuration file | ~233 |
-</claude-mem-context>
@@ -2,6 +2,7 @@ export interface Observation {
  id: number;
  memory_session_id: string;
  project: string;
+  merged_into_project?: string | null;
  platform_source: string;
  type: string;
  title: string | null;
@@ -1,58 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Nov 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #4035 | 10:24 PM | 🔵 | logger.ts file exists but is empty | ~220 |
-
-### Nov 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6521 | 5:43 PM | 🔵 | Code Review: Enhanced HTTP Logging and Double Entries Bug Fix | ~482 |
-
-### Nov 17, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #10019 | 12:14 AM | 🔵 | TranscriptParser Utility: JSONL Parsing with Type-Safe Entry Filtering | ~569 |
-
-### Nov 23, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #14626 | 6:25 PM | 🔵 | Stop Hook Summary Not in Transcript Validator Schema | ~359 |
-
-### Nov 28, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #17238 | 11:34 PM | 🔵 | Existing TranscriptParser TypeScript implementation handles nested message structure | ~493 |
-
-### Dec 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20407 | 7:20 PM | 🔵 | Tag stripping utilities implement dual-tag privacy system with ReDoS protection | ~415 |
-
-### Dec 8, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #22310 | 9:46 PM | 🟣 | Complete Hook Lifecycle Documentation Generated | ~603 |
-| #22306 | 9:45 PM | 🔵 | Dual-Tag Privacy System with ReDoS Protection | ~461 |
-
-### Dec 14, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #25691 | 4:24 PM | 🔵 | happy_path_error__with_fallback utility logs errors to silent.log and returns fallback values | ~460 |
-
-### Dec 20, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #30883 | 6:38 PM | 🔵 | Tag-Stripping DRY Violation Analysis | ~152 |
-</claude-mem-context>
@@ -58,21 +58,26 @@ export function getProjectName(cwd: string | null | undefined): string {
 * Project context with worktree awareness
 */
 export interface ProjectContext {
-  /** Canonical project name for writes/queries (parent repo in worktrees) */
+  /** Canonical project name for writes/queries; `parent/worktree` when in a worktree */
  primary: string;
  /** Parent project name if in a worktree, null otherwise */
  parent: string | null;
  /** True if currently in a worktree */
  isWorktree: boolean;
-  /** All projects to query: [primary] for main repo, [parentRepo, worktreeName] for worktree */
+  /** Projects to query for reads. In a worktree: `[parent, composite]` so
+   *  main-repo context flows into every worktree while sibling worktrees stay
+   *  isolated. In the main repo: `[primary]`. Writes always use `.primary`. */
  allProjects: string[];
 }

 /**
 * Get project context with worktree detection.
 *
- * When in a worktree, returns both the worktree project name and parent project name
- * for unified timeline queries.
+ * Each worktree is its own bucket. When in a worktree, `primary` is the
+ * composite `parent/worktree` (e.g. `claude-mem/dar-es-salaam`) so worktrees
+ * are uniquely identified and grouped under their parent project without
+ * mixing observations across them. In the main repo, `primary` is just the
+ * project basename.
 *
 * @param cwd - Current working directory (absolute path)
 * @returns ProjectContext with worktree info
@@ -88,14 +93,12 @@ export function getProjectContext(cwd: string | null | undefined): ProjectContex
  const worktreeInfo = detectWorktree(expandedCwd);

  if (worktreeInfo.isWorktree && worktreeInfo.parentProjectName) {
-    // In a worktree: use parent project name as primary so observations
-    // are stored under the same project as the main repo (#1081, #1500, #1819)
-    const allProjects = Array.from(new Set([worktreeInfo.parentProjectName, cwdProjectName]));
+    const composite = `${worktreeInfo.parentProjectName}/${cwdProjectName}`;
    return {
-      primary: worktreeInfo.parentProjectName,
+      primary: composite,
      parent: worktreeInfo.parentProjectName,
      isWorktree: true,
-      allProjects
+      allProjects: [worktreeInfo.parentProjectName, composite]
    };
  }

@@ -1,58 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Nov 10, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #6358 | 3:14 PM | 🔵 | SDK Agent Spatial Awareness Implementation | ~309 |
-
-### Nov 21, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #13289 | 2:20 PM | 🟣 | Comprehensive Test Suite for Transcript Transformation | ~320 |
-
-### Nov 23, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #14617 | 6:15 PM | 🟣 | Test Suite Successfully Passing - All 8 Tests Green | ~498 |
-| #14615 | 6:14 PM | 🟣 | YAGNI-Focused Test Suite for Transcript Transformation | ~457 |
-
-### Dec 5, 2025
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #20732 | 9:07 PM | 🔵 | Smart Install Version Marker Tests for Upgrade Detection | ~452 |
-| #20399 | 7:17 PM | 🔵 | Smart install tests validate version tracking with backward compatibility | ~311 |
-| #20392 | 7:15 PM | 🔵 | Memory tag stripping tests validate dual-tag system for JSON context filtering | ~404 |
-| #20391 | " | 🔵 | User prompt tag stripping tests validate privacy controls for memory exclusion | ~182 |
-
-### Jan 3, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36663 | 11:06 PM | ✅ | Third Validation Test Updated: Resume Safety Check Now Uses NULL Comparison | ~417 |
-| #36662 | " | ✅ | Second Validation Test Updated: Post-Capture Check Now Uses NULL Comparison | ~418 |
-| #36661 | 11:05 PM | ✅ | First Validation Test Updated: Placeholder Detection Now Checks for NULL | ~482 |
-| #36660 | " | ✅ | Updated Session ID Usage Validation Test Header to Reflect NULL-Based Architecture | ~588 |
-| #36659 | " | ✅ | Sixth Test Fix: Updated Multi-Observation Test to Use Memory Session ID | ~486 |
-| #36658 | " | ✅ | Fifth Test Fix: Updated storeSummary Tests to Use Actual Memory Session ID After Capture | ~555 |
-| #36657 | 11:04 PM | ✅ | Fourth Test Fix: Updated storeObservation Tests to Use Actual Memory Session ID After Capture | ~547 |
-| #36656 | " | ✅ | Third Test Fix: Updated getSessionById Test to Expect NULL for Uncaptured Memory Session ID | ~436 |
-| #36655 | " | ✅ | Second Test Fix: Updated updateMemorySessionId Test to Expect NULL Before Update | ~395 |
-| #36654 | " | ✅ | First Test Fix: Updated Memory Session ID Initialization Test to Expect NULL | ~426 |
-| #36650 | 11:02 PM | 🔵 | Phase 1 Analysis Reveals Implementation-Test Mismatch on NULL vs Placeholder Initialization | ~687 |
-| #36648 | " | 🔵 | Session ID Refactor Test Suite Documents Database Migration 17 and Dual ID System | ~651 |
-| #36647 | 11:01 PM | 🔵 | SessionStore Test Suite Validates Prompt Counting and Timestamp Override Features | ~506 |
-| #36646 | " | 🔵 | Session ID Architecture Revealed Through Test File Analysis | ~611 |
-
-### Jan 4, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36858 | 1:50 AM | 🟣 | Phase 1 Implementation Completed via Subagent | ~499 |
-| #36854 | 1:49 AM | 🟣 | gemini-3-flash Model Tests Added to GeminiAgent Test Suite | ~470 |
-| #36851 | " | 🔵 | GeminiAgent Test Structure Analyzed | ~565 |
-</claude-mem-context>
@@ -103,7 +103,7 @@ describe('AgentFormatter', () => {
      const result = renderAgentHeader('my-project');

      expect(result).toHaveLength(2);
-      expect(result[0]).toMatch(/^# \$CMEM my-project \d{4}-\d{2}-\d{2} \d{1,2}:\d{2}[ap]m [A-Z]{3,4}$/);
+      expect(result[0]).toMatch(/^# \[my-project\] recent context, \d{4}-\d{2}-\d{2} \d{1,2}:\d{2}[ap]m [A-Z]{3,4}$/);
      expect(result[1]).toBe('');
    });

@@ -116,7 +116,7 @@ describe('AgentFormatter', () => {
    it('should handle empty project name', () => {
      const result = renderAgentHeader('');

-      expect(result[0]).toMatch(/^# \$CMEM  \d{4}-\d{2}-\d{2} \d{1,2}:\d{2}[ap]m [A-Z]{3,4}$/);
+      expect(result[0]).toMatch(/^# \[\] recent context, \d{4}-\d{2}-\d{2} \d{1,2}:\d{2}[ap]m [A-Z]{3,4}$/);
    });
  });

@@ -452,7 +452,7 @@ describe('AgentFormatter', () => {
    it('should return helpful message with project name', () => {
      const result = renderAgentEmptyState('my-project');

-      expect(result).toContain('# $CMEM my-project');
+      expect(result).toContain('# [my-project] recent context,');
      expect(result).toContain('No previous sessions found.');
    });

@@ -466,7 +466,7 @@ describe('AgentFormatter', () => {
    it('should handle empty project name', () => {
      const result = renderAgentEmptyState('');

-      expect(result).toContain('# $CMEM ');
+      expect(result).toContain('# [] recent context,');
    });
  });
 });
@@ -1,13 +0,0 @@
-<claude-mem-context>
-# Recent Activity
-
-### Jan 4, 2026
-
-| ID | Time | T | Title | Read |
-|----|------|---|-------|------|
-| #36870 | 1:54 AM | 🟣 | Phase 2 Implementation Completed via Subagent | ~572 |
-| #36866 | 1:53 AM | 🔄 | WMIC Test Refactored to Use Direct Logic Testing | ~533 |
-| #36865 | 1:52 AM | ✅ | WMIC Test File Updated with Improved Mock Implementation | ~370 |
-| #36863 | 1:51 AM | 🟣 | WMIC Parsing Test File Created | ~581 |
-| #36861 | " | 🔵 | Existing ProcessManager Test File Structure Analyzed | ~516 |
-</claude-mem-context>
@@ -1,3 +0,0 @@
-<claude-mem-context>
-
-</claude-mem-context>
@@ -97,7 +97,7 @@ describe('getProjectContext', () => {
    expect(ctx.parent).toBeNull();
  });

-  describe('worktree regression (#1081, #1500, #1819)', () => {
+  describe('worktree isolation', () => {
    let tmp: string;
    let mainRepo: string;
    let worktreeCheckout: string;
@@ -125,21 +125,18 @@ describe('getProjectContext', () => {
      rmSync(tmp, { recursive: true, force: true });
    });

-    it('uses parent project name as primary when in a worktree', () => {
+    it('uses parent/worktree composite as primary when in a worktree', () => {
      const ctx = getProjectContext(worktreeCheckout);
      expect(ctx.isWorktree).toBe(true);
-      expect(ctx.primary).toBe('main-repo');
+      expect(ctx.primary).toBe('main-repo/my-worktree');
      expect(ctx.parent).toBe('main-repo');
-      expect(ctx.allProjects).toEqual(['main-repo', 'my-worktree']);
+      expect(ctx.allProjects).toEqual(['main-repo', 'main-repo/my-worktree']);
    });

-    it('write-path call sites resolve to parent project in worktrees', () => {
-      // Mirrors the pattern used by session-init.ts and SessionRoutes.ts:
-      //   const project = getProjectContext(cwd).primary;
-      // This must resolve to the parent repo, not the worktree name,
-      // so observations are stored under the correct project.
+    it('write-path call sites resolve to composite name in worktrees', () => {
      const project = getProjectContext(worktreeCheckout).primary;
-      expect(project).toBe('main-repo');
+      expect(project).toBe('main-repo/my-worktree');
+      expect(project).not.toBe('main-repo');
      expect(project).not.toBe('my-worktree');
    });
  });
@@ -319,9 +319,7 @@ describe('ResponseProcessor', () => {
      );

      const [, , , summary] = mockStoreObservations.mock.calls[0];
-      // #1718: When observations exist without <summary> tags, a synthetic summary is salvaged
-      expect(summary).not.toBeNull();
-      expect(summary.notes).toContain('Salvaged from');
+      expect(summary).toBeNull();
    });
  });
Author	SHA1	Message	Date
Alex Newman	11666e9ffb	chore: bump version to 12.2.0 Publish to npm / publish (push) Has been cancelled	2026-04-17 20:14:22 -07:00
Alex Newman	fb8e526c55	Merge pull request #2052 from thedotmack/thedotmack/worktree-remap feat(worktree): scope per worktree, cwd backfill, and merged-worktree adoption	2026-04-17 20:10:54 -07:00
Alex Newman	b8d63d949f	fix(worktree): self-heal Chroma metadata on re-run Addresses unresolved CodeRabbit finding on WorktreeAdoption.ts:296. Previously, Chroma patch failures stranded rows permanently: adoptedSqliteIds was built only from rows where merged_into_project IS NULL, so once SQL committed, reruns couldn't rediscover them for retry. The Chroma id set is now built from ALL observations whose project matches a merged worktree — including rows already stamped to this parent. Combined with the idempotent updateMergedIntoProject, transient Chroma failures self-heal on the next adoption pass. SQL writes remain idempotent (UPDATE still guards on merged_into_project IS NULL), so adoptedObservations / adoptedSummaries continue to count only newly-adopted rows. chromaUpdates now counts total Chroma writes per pass (may exceed adoptedObservations when retrying).	2026-04-16 22:01:21 -07:00
Alex Newman	7a66cb310f	fix(worktree): address PR review — schema guard, startup adoption, query parity Addresses six CodeRabbit/Greptile findings on PR #2052: - Schema guard in adoptMergedWorktrees probes for merged_into_project columns before preparing statements; returns early when absent so first boot after upgrade (pre-migration) doesn't silently fail. - Startup adoption now iterates distinct cwds from pending_messages and dedupes via resolveMainRepoPath — the worker daemon runs with cwd=plugin scripts dir, so process.cwd() fallback was a no-op. - ObservationCompiler single-project queries (queryObservations / querySummaries) OR merged_into_project into WHERE so injected context surfaces adopted worktree rows, matching the Multi variants. - SessionStore constructor now calls ensureMergedIntoProjectColumns so bundled artifacts (context-generator.cjs) that embed SessionStore get the merged_into_project column on DBs that only went through the bundled migration chain. - OBSERVER_SESSIONS_PROJECT constant is now derived from basename(OBSERVER_SESSIONS_DIR) and used across PaginationHelper, SessionStore, and timeline queries instead of hardcoded strings. - Corrected misleading Chroma retry docstring in WorktreeAdoption to match actual behavior (no auto-retry once SQL commits).	2026-04-16 21:31:30 -07:00
Alex Newman	d1601123fd	feat(ui): hide observer-sessions project from UI lists Observer sessions (internal SDK-driven worker queries) run under a synthetic project name 'observer-sessions' to keep them out of claude --resume. They were still surfacing in the viewer project picker and unfiltered observation/summary/prompt feeds. Filter them out at every UI-facing query: - SessionStore.getAllProjects and getProjectCatalog - timeline/queries.ts getAllProjects - PaginationHelper observations/summaries/prompts when no project is selected When a caller explicitly requests project='observer-sessions', results are still returned (not a hard ban, just hidden by default). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 20:05:37 -07:00
Alex Newman	f6fda8fff4	fix(worktree): address CodeRabbit PR review feedback - Document --branch override in npx-cli help text - Guard ContextBuilder against empty projects[] override; fall back to cwd-derived primary - Ensure merged_into_project indexes are created even if ALTER ran in a prior partial migration - Reject adopt --branch/--cwd flags with missing or flag-like values - Use defined --color-border-primary token for merged badge border Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 20:03:27 -07:00
Alex Newman	d24f3a7019	fix(worktree): address PR review — test assertion, dry-run sentinel, git timeouts - Update allProjects test expectation to match [parent, composite] (matches JSDoc + callers in ContextBuilder/context handlers). - Replace string-matched __DRY_RUN_ROLLBACK__ sentinel with dedicated DryRunRollback class to avoid swallowing unrelated errors. - Add 5000ms timeout to spawnSync git calls in WorktreeAdoption and ProcessManager so worker startup can't hang on a stuck git process. - Drop unreachable break after process.exit(0) in adopt case. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:50:01 -07:00
Alex Newman	0a5f072aaf	build(worktree): rebuild plugin artifacts for worktree adoption feature Regenerated worker-service.cjs, context-generator.cjs, viewer.html, and viewer-bundle.js to reflect all six implementation phases of the merged- worktree adoption feature. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:36:00 -07:00
Alex Newman	bce4ce32ec	feat(ui): show merged-into-parent badge on adopted observations ObservationCard renders a secondary "merged → <parent>" chip when merged_into_project is set, next to the existing project label. Both are meaningful: project is origin provenance, merged_into_project is the current home. Extends PaginationHelper's observations and summaries queries with OR merged_into_project = ? so the single-project viewer fetch pulls in adopted rows — the plan's Phase 3 covered multi-project context injection; this is the single-project UI read path. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:33:47 -07:00
Alex Newman	5664fabce4	feat(cli): npx claude-mem adopt [--dry-run] [--branch X] Adds a manual escape hatch for the worktree adoption engine. Covers squash-merges where git branch --merged HEAD returns nothing, and lets users re-run adoption on demand. Wired through worker-service.cjs (same pattern as generate/clean) so the command runs under Bun with bun:sqlite, keeping npx-cli/ pure Node. --cwd flag passes the user's working directory through the spawn so the engine resolves the correct parent repo. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:28:17 -07:00
Alex Newman	0b90495391	feat(worktree): auto-adopt merged worktrees on worker startup Invokes adoptMergedWorktrees() right after runOneTimeCwdRemap() and before dbManager.initialize(), wrapped in try/catch so adoption failures never block startup. Idempotent, so running every startup is cheap — the SQL UPDATE only touches rows where merged_into_project IS NULL. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:24:43 -07:00
Alex Newman	3e770df332	feat(worktree): query plumbing surfaces merged rows under parent project ObservationCompiler.queryObservationsMulti and querySummariesMulti WHERE clause extended with OR merged_into_project IN (...), so a parent-project read pulls in rows originally written under any child worktree's composite name once merged. SearchManager wraps the Chroma project filter in \$or so semantic search behaves identically. ChromaSync baseMetadata now carries merged_into_project on new embeddings; existing rows are patched retroactively by the adoption engine. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:24:10 -07:00
Alex Newman	a7c3c4af2d	feat(worktree): adoption engine for merged worktrees Detects merged worktrees via git (worktree list --porcelain + branch --merged HEAD), then stamps merged_into_project on SQLite observations/summaries and propagates the same metadata to Chroma in lockstep. `project` stays immutable; adoption is a virtual pointer. Idempotent via IS NULL guard on UPDATE and by idempotent Chroma metadata writes. SQL is source of truth — Chroma failures are logged but don't roll back SQL. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:19:02 -07:00
Alex Newman	3d1dfcc26a	feat(migration): add merged_into_project column for worktree adoption Nullable pointer on observations and session_summaries that lets a worktree's rows surface under the parent project's observation list without data movement. Self-idempotent via PRAGMA table_info guard; does not bump schema_versions. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 19:12:38 -07:00
Alex Newman	dc198d5677	feat(worktree): include parent repo observations in worktree read scope Worktrees are branches off main; the parent holds the architecture, decisions, and long-tail history the worktree inherits. Scoping reads to the worktree alone meant every new worktree started cold on any question that required prior context. Expand `allProjects` in a worktree to `[parent, composite]` so reads pull both. Writes still go through `.primary` (the composite), so sibling worktrees don't leak into each other. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 17:55:17 -07:00
Alex Newman	193e7e0719	feat(worktree): auto-apply cwd-based project remap on worker startup Ports scripts/cwd-remap.ts into ProcessManager.runOneTimeCwdRemap() and invokes it in initializeBackground() alongside the existing chroma migration. Uses pending_messages.cwd as the source of truth to rewrite pre-worktree bare project names into the parent/worktree composite format so search and context are consistent. - Backs up the DB to .bak-cwd-remap-<ts> before any writes. - Idempotent: marker file .cwd-remap-applied-v1 short-circuits reruns. - No-ops on fresh installs (no DB, or no pending_messages table). - On failure, logs and skips the marker so the next restart retries. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 17:51:33 -07:00
Alex Newman	9d695f53ed	chore: remove auto-generated per-directory CLAUDE.md files Leftover artifacts from an abandoned context-injection feature. The project-level CLAUDE.md stays; the directory-level ones were generated timeline scaffolding that never panned out. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 17:51:24 -07:00
Alex Newman	a65ab055ca	fix(worktree): audit observation fetch/display for composite project names Three sites didn't account for parent/worktree composite naming: - PaginationHelper.stripProjectPath: marker used full composite, breaking path sanitization for worktrees checked out outside a parent/leaf layout. Now extracts the leaf segment. - observations/store.ts: fallback imported getCurrentProjectName from shared/paths.ts (a duplicate impl without worktree detection). Switched to getProjectContext().primary so writes key into the same project as reads. - SearchManager.getRecentContext: fallback used basename(cwd) and lost the parent prefix, making the MCP tool find nothing in worktrees. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 17:38:49 -07:00
Alex Newman	d589bc5f25	fix(cwd-remap): address PR review feedback - Handle bare repo common-dir (strip trailing .git) instead of an identical-branch ternary - Surface unexpected git stderr while keeping "not a git repository" silent - Explicitly close the sqlite handle in both dry-run and apply paths so WAL checkpoints complete Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 17:21:49 -07:00
Alex Newman	3869b083d0	fix(context): derive project from explicit projects array, not cwd When a caller (e.g. worker context-inject route) passes a `projects` array without a matching cwd, the cwd-derived `context.primary` drifted from the projects being queried — producing an empty-state header for one project while querying another. Use the last entry of `projects` so header and query target stay in sync.	2026-04-16 17:16:51 -07:00
Alex Newman	148e1892df	chore(scripts): replace worktree-remap with cwd-based remap using pending_messages.cwd The old worktree-remap.ts tried to reconstruct per-session cwd by regex- matching absolute paths that incidentally leaked into observation free-text (files_read, source_input_summary, metadata, user_prompt). That source is derived and lossy: it only hit 1/3498 plain-project sessions in practice. pending_messages.cwd is the structured, authoritative cwd captured from every hook payload — 7,935 of 8,473 rows are populated. cwd-remap.ts uses that column as the source of truth: 1. Pull every distinct cwd from pending_messages.cwd 2. For each cwd, classify with git: - rev-parse --absolute-git-dir vs --git-common-dir → main vs worktree - rev-parse --show-toplevel for the correct leaf (handles cwds that are subdirs of the worktree root) Parent project name = basename(dirname(common-dir)); composite is parent/worktree for worktrees, basename(toplevel) for main repos. 3. For each session, take the EARLIEST pending_messages.cwd (not the dominant one — claude-mem's own hooks run from nested .context/ claude-mem/ directories and would otherwise poison the count). 4. Apply UPDATEs in a single transaction across sdk_sessions, observations, and session_summaries. Auto-backs-up the DB first. Result on a real DB: 41 sessions remapped (vs 1 previously), 1,694 observations and 3,091 session_summaries updated to match. 43 cwds skipped (deleted worktrees / non-repos) are left untouched — no inference when the data isn't there. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>	2026-04-16 16:40:37 -07:00
Alex Newman	040729beef	fix(project-name): use parent/worktree composite so observations don't cross worktrees Revert of #1820 behavior. Each worktree now gets its own bucket: - In a worktree, primary = `parent/worktree` (e.g. `claude-mem/dar-es-salaam`) - In a main repo, primary = basename (unchanged) - allProjects is always `[primary]` — strict isolation at query time Includes a one-off maintenance script (scripts/worktree-remap.ts) that retroactively reattributes past sessions to their worktree using path signals in observations and user prompts. Two-rule inference keeps the remap high-confidence: 1. The worktree basename in the path matches the session's current plain project name (pre-#1820 era; trusted). 2. Or all worktree path signals converge on a single (parent, worktree) across the session. Ambiguous sessions are skipped. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>	2026-04-16 15:40:44 -07:00
Alex Newman	53622b59e9	docs: update CHANGELOG.md for v12.1.6 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 14:35:40 -07:00
Alex Newman	69080dc291	chore: bump version to 12.1.6 Publish to npm / publish (push) Has been cancelled Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 14:31:43 -07:00
Alex Newman	c76a439491	fix: drop orphan flag when filtering empty-string spawn args (#2049 ) Observations were 100% failing on Claude Code 2.1.109+ because the Agent SDK emits ["--setting-sources", ""] when settingSources defaults to []. The existing Bun-workaround filter stripped the empty string but left the orphan --setting-sources flag, which then consumed --permission-mode as its value, crashing the subprocess with: Error processing --setting-sources: Invalid setting source: --permission-mode. Make the filter pair-aware: when an empty arg follows a --flag, drop both so the SDK default (no setting sources) is preserved by omission. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-04-16 14:30:54 -07:00
Alex Newman	70a150db74	docs: update CHANGELOG.md for v12.1.5	2026-04-15 14:41:31 -07:00
Alex Newman	d7b4610e27	chore: bump version to 12.1.5 Publish to npm / publish (push) Has been cancelled	2026-04-15 14:40:44 -07:00
Alex Newman	88bb4e589e	docs: update CHANGELOG.md for v12.1.4 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-15 12:10:29 -07:00
Alex Newman	ebefae864e	chore: bump version to 12.1.4 Publish to npm / publish (push) Has been cancelled Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-15 12:06:52 -07:00
Alex Newman	0cd931bb06	Merge pull request #1865 from thedotmack/thedotmack/find-cmem-refs fix: revert unauthorized $CMEM branding in context header	2026-04-15 12:06:10 -07:00
Alex Newman	4c792f026d	build: rebuild plugin artifacts after $CMEM header revert Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-15 12:05:49 -07:00
Alex Newman	aa7cdb6d9f	fix: revert unauthorized $CMEM branding in context header A prior Claude instance snuck in a `$CMEM` token branding header during a context compression refactor (`7e072106`). Reverts back to the original descriptive format: `# [project] recent context, datetime` Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-15 12:04:27 -07:00
Alex Newman	5db90f2ea0	docs: update CHANGELOG.md for v12.1.3	2026-04-15 11:43:49 -07:00
Alex Newman	4ddf57610a	chore: bump version to 12.1.3 Publish to npm / publish (push) Has been cancelled	2026-04-15 04:26:29 -07:00
Alex Newman	d0fc68c630	revert: remove overengineered summary salvage logic (#1718 ) (#1850 ) The synthetic summary salvage feature created fake summaries from observation data when the AI returned <observation> instead of <summary> tags. This was overengineered — missing a summary is preferable to fabricating one from observation fields that don't map cleanly to summary semantics. Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-04-15 04:22:41 -07:00
				`@@ -1 +0,0 @@`
				Never read built source files in this directory. These are compiled outputs — read the source files in `src/` instead.