* docs: add folder index generator plan RFC for auto-generating folder-level CLAUDE.md files with observation timelines. Includes IDE symlink support and root CLAUDE.md integration. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: implement folder index generator (Phase 1) Add automatic CLAUDE.md generation for folders containing observed files. This enables IDE context providers to access relevant memory observations. Core modules: - FolderDiscovery: Extract folders from observation file paths - FolderTimelineCompiler: Compile chronological timeline per folder - ClaudeMdGenerator: Write CLAUDE.md with tag-based content replacement - FolderIndexOrchestrator: Coordinate regeneration on observation save Integration: - Event-driven regeneration after observation save in ResponseProcessor - HTTP endpoints for folder discovery, timeline, and manual generation - Settings for enabling/configuring folder index behavior The <claude-mem-context> tag wrapping ensures: - Manual CLAUDE.md content is preserved - Auto-generated content won't be recursively observed - Clean separation between user and system content 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: add updateFolderClaudeMd function to CursorHooksInstaller Adds function to update CLAUDE.md files for folders touched by observations. Uses existing /api/search/by-file endpoint, preserves content outside <claude-mem-context> tags, and writes atomically via temp file + rename. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: hook updateFolderClaudeMd into ResponseProcessor Calls updateFolderClaudeMd after observation save to update folder-level CLAUDE.md files. Uses fire-and-forget pattern with error logging. Extracts file paths from saved observations and workspace path from registry. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: add timeline formatting for folder CLAUDE.md files Implements formatTimelineForClaudeMd function that transforms API response into compact markdown table format. Converts emojis to text labels, handles ditto marks for timestamps, and groups under "Recent" header. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * refactor: remove old folder-index implementation Deletes redundant folder-index services that were replaced by the simpler updateFolderClaudeMd approach in CursorHooksInstaller.ts. Removed: - src/services/folder-index/ directory (5 files) - FolderIndexRoutes.ts - folder-index settings from SettingsDefaultsManager - folder-index route registration from worker-service 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: add worktree-aware project filtering for unified timelines Detect git worktrees and show both parent repo and worktree observations in the session start timeline. When running in a worktree, the context now includes observations from both projects, interleaved chronologically. - Add detectWorktree() utility to identify worktree directories - Add getProjectContext() to return parent + worktree projects - Update context hook to pass multi-project queries - Add queryObservationsMulti() and querySummariesMulti() for IN clauses - Maintain backward compatibility with single-project queries 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> * fix: restructure logging to prove session correctness and reduce noise Add critical logging at each stage of the session lifecycle to prove the session ID chain (contentSessionId → sessionDbId → memorySessionId) stays aligned. New logs include CREATED, ENQUEUED, CLAIMED, MEMORY_ID_CAPTURED, STORING, and STORED. Move intermediate migration and backfill progress logs to DEBUG level to reduce noise, keeping only essential initialization and completion logs at INFO level. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com> * refactor: extract folder CLAUDE.md utils to shared location Moves folder CLAUDE.md utilities from CursorHooksInstaller to a new shared utils file. Removes Cursor registry dependency - file paths from observations are already absolute, no workspace lookup needed. New file: src/utils/claude-md-utils.ts - replaceTaggedContent() - preserves user content outside tags - writeClaudeMdToFolder() - atomic writes with tag preservation - formatTimelineForClaudeMd() - API response to compact markdown - updateFolderClaudeMdFiles() - orchestrates folder updates 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix: trigger folder CLAUDE.md updates when observations are saved The folder CLAUDE.md update was previously only triggered in syncAndBroadcastSummary, but summaries run with observationCount=0 (observations are saved separately). Moved the update logic to syncAndBroadcastObservations where file paths are available. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * all the claudes * test: add unit tests for claude-md-utils pure functions Add 11 tests covering replaceTaggedContent and formatTimelineForClaudeMd: - replaceTaggedContent: empty content, tag replacement, appending, partial tags - formatTimelineForClaudeMd: empty input, parsing, ditto marks, session IDs 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * test: add integration tests for file operation functions Add 9 tests for writeClaudeMdToFolder and updateFolderClaudeMdFiles: - writeClaudeMdToFolder: folder creation, content preservation, nested dirs, atomic writes - updateFolderClaudeMdFiles: empty skip, fetch/write, deduplication, error handling 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * test: add unit tests for timeline-formatting utilities Add 14 tests for extractFirstFile and groupByDate functions: - extractFirstFile: relative paths, fallback to files_read, null handling, invalid JSON - groupByDate: empty arrays, date grouping, chronological sorting, item preservation 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * chore: rebuild plugin scripts with merged features * docs: add project-specific CLAUDE.md with architecture and development notes * fix: exclude project root from auto-generated CLAUDE.md updates Skip folders containing .git directory when auto-updating subfolder CLAUDE.md files. This ensures: 1. Root CLAUDE.md remains user-managed and untouched by the system 2. SessionStart context injection stays pristine throughout the session 3. Subfolder CLAUDE.md files continue to receive live context updates 4. Cleaner separation between user-authored root docs and auto-generated folder indexes 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix: prevent crash from resuming stale SDK sessions on worker restart When the worker restarts, it was incorrectly passing the `resume` parameter to INIT prompts (lastPromptNumber=1) when a memorySessionId existed from a previous SDK session. This caused "Claude Code process exited with code 1" crashes because the SDK tried to resume into a session that no longer exists. Root cause: The resume condition only checked `hasRealMemorySessionId` but did not verify that this was a CONTINUATION prompt (lastPromptNumber > 1). Fix: Add `session.lastPromptNumber > 1` check to the resume condition: - Before: `...(hasRealMemorySessionId && { resume: session.memorySessionId })` - After: `...(hasRealMemorySessionId && session.lastPromptNumber > 1 && { resume: ... })` Also added: - Enhanced debug logging that warns when skipping resume for INIT prompts - Unit tests in tests/sdk-agent-resume.test.ts (9 test cases) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix: properly handle Chroma MCP connection errors Previously, ensureCollection() caught ALL errors from chroma_get_collection_info and assumed they meant "collection doesn't exist", triggering unnecessary collection creation attempts. Connection errors like "Not connected" or "MCP error -32000: Connection closed" would cascade into failed creation attempts. Similarly, queryChroma() would silently return empty results when the MCP call failed, masking the underlying connection problem. Changes: - ensureCollection(): Detect connection errors and re-throw immediately instead of attempting collection creation - queryChroma(): Wrap MCP call in try-catch and throw connection errors instead of returning empty results - Both methods reset connection state (connected=false, client=null) on connection errors so subsequent operations can attempt to reconnect 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * pushed * fix: scope regenerate-claude-md.ts to current working directory Critical bug fix: The script was querying ALL observations from the database across ALL projects ever recorded (1396+ folders), then attempting to write CLAUDE.md files everywhere including other projects, non-existent paths, and ignored directories. Changes: - Use git ls-files to discover folders (respects .gitignore automatically) - Filter database query to current project only (by folder name) - Use relative paths for database queries (matches storage format) - Add --clean flag to remove auto-generated CLAUDE.md files - Add fallback directory walker for non-git repos Now correctly scopes to 26 folders with observations instead of 1396+. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * docs and adjustments * fix: cleanup mode strips tags instead of deleting files blindly The cleanup mode was incorrectly deleting entire files that contained <claude-mem-context> tags. The correct behavior (per original design): 1. Strip the <claude-mem-context>...</claude-mem-context> section 2. If empty after stripping → delete the file 3. If has remaining content → save the stripped version Now properly preserves user content in CLAUDE.md files while removing only the auto-generated sections. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * deleted some files * chore: regenerate folder CLAUDE.md files with fixed script Regenerated 23 folder CLAUDE.md files using the corrected script that: - Scopes to current working directory only - Uses git ls-files to respect .gitignore - Filters by project name 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * Update CLAUDE.md files for January 5, 2026 - Regenerated and staged 23 CLAUDE.md files with a mix of new and modified content. - Fixed cleanup mode to properly strip tags instead of deleting files blindly. - Cleaned up empty CLAUDE.md files from various directories, including ~/.claude and ~/Scripts. - Conducted dry-run cleanup that identified a significant reduction in auto-generated CLAUDE.md files. - Removed the isAutoGeneratedClaudeMd function due to incorrect file deletion behavior. * feat: use settings for observation limit in batch regeneration script Replace hard-coded limit of 10 with configurable CLAUDE_MEM_CONTEXT_OBSERVATIONS setting (default: 50). This allows users to control how many observations appear in folder CLAUDE.md files. Changes: - Import SettingsDefaultsManager and load settings at script startup - Use OBSERVATION_LIMIT constant derived from settings at both call sites - Remove stale default parameter from findObservationsByFolder function 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: use settings for observation limit in event-driven updates Replace hard-coded limit of 10 in updateFolderClaudeMdFiles with configurable CLAUDE_MEM_CONTEXT_OBSERVATIONS setting (default: 50). Changes: - Import SettingsDefaultsManager and os module - Load settings at function start (once, not in loop) - Use limit from settings in API call 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: Implement configurable observation limits and enhance search functionality - Added configurable observation limits to batch regeneration scripts. - Enhanced SearchManager to handle folder queries and normalize parameters. - Introduced methods to check for direct child files in observations and sessions. - Updated SearchOptions interface to include isFolder flag for filtering. - Improved code quality with comprehensive reviews and anti-pattern checks. - Cleaned up auto-generated CLAUDE.md files across various directories. - Documented recent changes and improvements in CLAUDE.md files. * build asset * Project Context from Claude-Mem auto-added (can be auto removed at any time) * CLAUDE.md updates * fix: resolve CLAUDE.md files to correct directory in worktree setups When using git worktrees, CLAUDE.md files were being written relative to the worker's process.cwd() instead of the actual project directory. This fix threads the project's cwd from message processing through to the file writing utilities, ensuring CLAUDE.md files are created in the correct project directory regardless of where the worker was started. Changes: - Add projectRoot parameter to updateFolderClaudeMdFiles for path resolution - Thread projectRoot through ResponseProcessor call chain - Track lastCwd from messages in SDKAgent, GeminiAgent, OpenRouterAgent - Add tests for relative/absolute path handling with projectRoot 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * more project context updates * context updates * fix: preserve actual dates in folder CLAUDE.md generation Previously, formatTimelineForClaudeMd used today's date for all observations because the API only returned time (e.g., "4:30 PM") without date information. This caused all historical observations to appear as if they happened today. Changes: - SearchManager.findByFile now groups results by date with headers (e.g., "### Jan 4, 2026") matching formatSearchResults behavior - formatTimelineForClaudeMd now parses these date headers and uses the correct date when constructing epochs for date grouping The timeline dates are critical for claude-mem context - LLMs need accurate temporal context to understand when work happened. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * build: update worker assets with date parsing fix 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * claude-mem context: Fixed critical date parsing bug in PR #556 * fix: address PR #556 review items - Use getWorkerHost() instead of hard-coded 127.0.0.1 in claude-md-utils - Add error message and stack details to FOLDER_INDEX logging - Add 5 new tests for worktree/projectRoot path resolution 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * Refactor CLAUDE documentation across multiple components and tests - Updated CLAUDE.md files in src/ui/viewer, src/ui/viewer/constants, src/ui/viewer/hooks, tests/server, tests/worker/agents, and plans to reflect recent changes and improvements. - Removed outdated entries and consolidated recent activities for clarity. - Enhanced documentation for hooks, settings, and pagination implementations. - Streamlined test suite documentation for server and worker agents, indicating recent test audits and cleanup efforts. - Adjusted plans to remove obsolete entries and focus on current implementation strategies. * docs: comprehensive v9.0 documentation audit and updates - Add usage/folder-context to docs.json navigation (was documented but hidden!) - Update introduction.mdx with v9.0 release notes (Live Context, Worktree Support, Windows Fixes) - Add CLAUDE_MEM_WORKER_HOST setting to configuration.mdx - Add Folder Context Files section with link to detailed docs - Document worktree support in folder-context.mdx - Update terminology from "mem-search skill" to "MCP tools" throughout active docs - Update Search Pipeline in architecture/overview.mdx - Update usage/getting-started.mdx with MCP tools terminology - Update usage/claude-desktop.mdx title and terminology - Update hooks-architecture.mdx reference 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: add recent activity log for worker CLI with detailed entries * chore: update CLAUDE.md context files 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * feat: add brainstorming report for CLAUDE.md distribution architecture --------- Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
9.0 KiB
Folder CLAUDE.md Generator
CORE DIRECTIVE (NON-NEGOTIABLE)
EXTEND THE EXISTING CURSOR RULES TIMELINE GENERATION SYSTEM TO ALSO WRITE CLAUDE.MD FILES
- DO NOT create new services
- DO NOT create new orchestrators
- DO NOT create new HTTP routes
- DO NOT create new database query functions
- EXTEND existing functions to add folder-level output
Approved Directives (From Planning Conversation)
Trigger Mechanism
- Observation save triggers folder CLAUDE.md regeneration INLINE
- NO batching
- NO debouncing
- NO Set-based queuing
- NO session-end hook
- Synchronous:
observation.save()→ update folder CLAUDE.md files → done
Tag Strategy
- Wrap ONLY auto-generated content with
<claude-mem-context>tags - Everything outside tags is untouched (user's manual content preserved)
- If tags are deleted, just regenerate them
- NO backup system
- NO manual content markers
Git Behavior
- CLAUDE.md files SHOULD be committed (intentional)
<claude-mem-context>tag is searchable fingerprint for GitHub analytics- NO .gitignore for these files
Phasing
- Phase 1: CLAUDE.md generation only (THIS PLAN)
- Phase 2: IDE symlinks (FUTURE)
REJECTED
- Cross-folder linking — NO
- Semantic grouping — deferred enhancement only
- Team sync — future phase
DEFERRED
- Priority weighting by observation type
- IDE-specific template refinements
Phase 0: Documentation Discovery (COMPLETED)
Existing APIs to USE (Not Rebuild)
| Function | Location | Purpose |
|---|---|---|
findByFile(filePath, options) |
src/services/sqlite/SessionSearch.ts:342 |
Query observations by folder prefix (already supports LIKE wildcards) |
updateCursorContextForProject() |
src/services/integrations/CursorHooksInstaller.ts:98 |
Write context files after observation save |
writeContextFile() |
src/utils/cursor-utils.ts:97 |
Atomic file write with temp file + rename |
extractFirstFile() |
src/shared/timeline-formatting.ts |
Extract file paths from JSON arrays |
groupByDate() |
src/shared/timeline-formatting.ts |
Group items chronologically |
formatTime(), formatDate() |
src/shared/timeline-formatting.ts |
Time formatting |
Existing Integration Points
| Location | What Happens | Extension Point |
|---|---|---|
ResponseProcessor.ts:266 |
Calls updateCursorContextForProject() after summary save |
Add folder CLAUDE.md update here |
CursorHooksInstaller.ts:98 |
updateCursorContextForProject() fetches context and writes file |
Add sibling function for folder updates |
Anti-Patterns to AVOID
- Creating
FolderIndexOrchestrator.ts— NO - Creating
FolderTimelineCompiler.ts— NO - Creating
FolderDiscovery.ts— NO - Creating
ClaudeMdGenerator.ts— NO - Creating
FolderIndexRoutes.ts— NO - Adding new HTTP endpoints — NO
- Adding new settings in
SettingsDefaultsManager.ts— NO (use sensible defaults inline)
Phase 1: Extend CursorHooksInstaller
What to Implement
Add ONE new function to src/services/integrations/CursorHooksInstaller.ts:
/**
* Update CLAUDE.md files for folders touched by an observation.
* Called inline after observation save, similar to updateCursorContextForProject.
*/
export async function updateFolderClaudeMd(
workspacePath: string,
filesModified: string[],
filesRead: string[],
project: string,
port: number
): Promise<void>
Implementation Pattern (Copy From)
Follow the EXACT pattern of updateCursorContextForProject() at line 98:
- Extract unique folder paths from filesModified and filesRead
- For each folder, fetch timeline via existing
/api/search/file?files=<folderPath>endpoint - Format as simple timeline (reuse existing formatters)
- Write to
<folder>/CLAUDE.mdpreserving content outside<claude-mem-context>tags
Tag Preservation Logic
function replaceTaggedContent(existingContent: string, newContent: string): string {
const startTag = '<claude-mem-context>';
const endTag = '</claude-mem-context>';
// If no existing content, wrap new content in tags
if (!existingContent) {
return `${startTag}\n${newContent}\n${endTag}`;
}
// If existing has tags, replace only tagged section
const startIdx = existingContent.indexOf(startTag);
const endIdx = existingContent.indexOf(endTag);
if (startIdx !== -1 && endIdx !== -1) {
return existingContent.substring(0, startIdx) +
`${startTag}\n${newContent}\n${endTag}` +
existingContent.substring(endIdx + endTag.length);
}
// If no tags exist, append tagged content at end
return existingContent + `\n\n${startTag}\n${newContent}\n${endTag}`;
}
Verification Checklist
- Function added to CursorHooksInstaller.ts
- Uses existing
findByFileendpoint (no new database queries) - Preserves content outside
<claude-mem-context>tags - Atomic writes (temp file + rename)
- Build passes:
npm run build
Phase 2: Hook Into ResponseProcessor
What to Implement
Add call to updateFolderClaudeMd() in src/services/worker/agents/ResponseProcessor.ts, right after the existing updateCursorContextForProject() call at line 266.
Code Location
In syncAndBroadcastSummary() function, after line 269:
// EXISTING: Update Cursor context file for registered projects (fire-and-forget)
updateCursorContextForProject(session.project, getWorkerPort()).catch(error => {
logger.warn('CURSOR', 'Context update failed (non-critical)', { project: session.project }, error as Error);
});
// NEW: Update folder CLAUDE.md files for touched folders (fire-and-forget)
// Extract file paths from the saved observations
updateFolderClaudeMd(
workspacePath, // From registry lookup
filesModified, // From observations
filesRead, // From observations
session.project,
getWorkerPort()
).catch(error => {
logger.warn('FOLDER_INDEX', 'CLAUDE.md update failed (non-critical)', { project: session.project }, error as Error);
});
Data Flow
processAgentResponse()saves observations → gets backobservationIds- Fetch observation records to get
files_readandfiles_modified - Pass to
updateFolderClaudeMd()
Verification Checklist
- Call added to ResponseProcessor.ts
- Fire-and-forget pattern (non-blocking, errors logged)
- Uses existing observation data (no new queries)
- Build passes:
npm run build
Phase 3: Timeline Formatting
What to Implement
Create a minimal timeline formatter for CLAUDE.md output. This can be:
- A simple function in CursorHooksInstaller.ts, OR
- Reuse existing
ResultFormatter.formatSearchResults()fromsrc/services/worker/search/ResultFormatter.ts
Output Format
# Recent Activity
<!-- This section is auto-generated by claude-mem. Edit content outside the tags. -->
<claude-mem-context>
### 2026-01-04
| Time | Type | Title |
|------|------|-------|
| 4:30pm | feature | Added folder index support |
| 3:15pm | bugfix | Fixed file path handling |
### 2026-01-03
| Time | Type | Title |
|------|------|-------|
| 11:00am | refactor | Cleaned up cursor utils |
</claude-mem-context>
Key Points
- Compact format (time, type emoji, title only)
- Grouped by date
- Limited to last N days or observations (sensible default: 10)
- NO token counts
- NO file columns (redundant - we're IN the folder)
Verification Checklist
- Formatter produces clean markdown
- Output is concise (not verbose)
- Grouped by date
- Build passes:
npm run build
Phase 4: Verification
Functional Tests
-
Manual Test:
- Start worker:
npm run dev - Create a test observation touching
src/services/sqlite/ - Verify
src/services/sqlite/CLAUDE.mdis created/updated - Verify
<claude-mem-context>tags are present - Verify manual content outside tags is preserved
- Start worker:
-
Build Check:
npm run build -
Grep for Anti-Patterns:
# Should find NOTHING grep -r "FolderIndexOrchestrator" src/ grep -r "FolderTimelineCompiler" src/ grep -r "FolderDiscovery" src/ grep -r "ClaudeMdGenerator" src/ grep -r "FolderIndexRoutes" src/ -
Grep for Correct Implementation:
# Should find the new function grep -r "updateFolderClaudeMd" src/
Tag Preservation Test
-
Create
src/test-folder/CLAUDE.mdwith manual content:# My Notes This is manual content I wrote. -
Trigger observation save touching files in
src/test-folder/ -
Verify result:
# My Notes This is manual content I wrote. <claude-mem-context> ### 2026-01-04 | Time | Type | Title | ... </claude-mem-context>
Summary
This is a ~100 line change spread across 2 files:
CursorHooksInstaller.ts— AddupdateFolderClaudeMd()function (~60 lines)ResponseProcessor.ts— Add call to the new function (~10 lines)
NO new files. NO new services. NO new routes. Just extending existing patterns.