v10.0.8

Bug Fixes

Orphaned Subprocess Cleanup

• Add explicit subprocess cleanup after SDK query loop using existing ProcessRegistry infrastructure (getProcessBySession + ensureProcessExit), preventing orphaned Claude subprocesses from accumulating
• Closes #1010, #1089, #1090, #1068

Chroma Binary Resolution

• Replace npx chroma run with absolute binary path resolution via require.resolve, falling back to npx with explicit cwd when the binary isn't found directly
• Closes #1120

Cross-Platform Embedding Fix

• Remove @chroma-core/default-embed which pulled in onnxruntime + sharp native binaries that fail on many platforms
• Use WASM backend for Chroma embeddings, eliminating native binary compilation issues
• Closes #1104, #1105, #1110

v10.0.7

Chroma HTTP Server Architecture

• Persistent HTTP server: Switched from in-process Chroma to a persistent HTTP server managed by the new ChromaServerManager for better reliability and performance
• Local embeddings: Added DefaultEmbeddingFunction for local vector embeddings — no external API required
• Pinned chromadb v3.2.2: Fixed compatibility with v2 API heartbeat endpoint
• Server lifecycle improvements: Addressed PR review feedback for proper start/stop/health check handling

Bug Fixes

• Fixed SDK spawn failures and sharp native binary crashes
• Added plugin.json to root .claude-plugin directory for proper plugin structure
• Removed duplicate else block from merge artifact

Infrastructure

• Added multi-tenancy support for claude-mem Pro
• Updated OpenClaw install URLs to install.cmem.ai
• Added Vercel deploy workflow for install scripts
• Added .claude/plans and .claude/worktrees to .gitignore

v10.0.6

Bug Fixes

• OpenClaw: Fix MEMORY.md project query mismatch — syncMemoryToWorkspace now includes both the base project name and the agent-scoped project name (e.g., both "openclaw" and "openclaw-main") when querying for context injection, ensuring the correct observations are pulled into MEMORY.md.

• OpenClaw: Add feed botToken support for Telegram — Feeds can now configure a dedicated botToken for direct Telegram message delivery, bypassing the OpenClaw gateway channel. This fixes scenarios where the gateway bot token couldn't be used for feed messages.

Other

• Changed OpenClaw plugin kind from "integration" to "memory" for accuracy.

v10.0.5

OpenClaw Installer & Distribution

This release introduces the OpenClaw one-liner installer and fixes several OpenClaw plugin issues.

New Features

• OpenClaw Installer (openclaw/install.sh): Full cross-platform installer script with curl | bash support
  • Platform detection (macOS, Linux, WSL)
  • Automatic dependency management (Bun, uv, Node.js)
  • Interactive AI provider setup with settings writer
  • OpenClaw gateway detection, plugin install, and memory slot configuration
  • Worker startup and health verification with rich diagnostics
  • TTY detection, --provider/--api-key CLI flags
  • Error recovery and upgrade handling for existing installations
  • jq/python3/node fallback chain for JSON config writing
• Distribution readiness tests (openclaw/test-install.sh): Comprehensive test suite for the installer
• Enhanced /api/health endpoint: Now returns version, uptime, workerPath, and AI status

Bug Fixes

• Fix: use event.prompt instead of ctx.sessionKey for prompt storage in OpenClaw plugin
• Fix: detect both openclaw and openclaw.mjs binary names in gateway discovery
• Fix: pass file paths via env vars instead of bash interpolation in node -e calls
• Fix: handle stale plugin config that blocks OpenClaw CLI during reinstall
• Fix: remove stale memory slot reference during reinstall cleanup
• Fix: remove opinionated filters from OpenClaw plugin

v10.0.4

Revert: v10.0.3 chroma-mcp spawn storm fix

v10.0.3 introduced regressions. This release reverts the codebase to the stable v10.0.2 state.

What was reverted

• Connection mutex via promise memoization
• Pre-spawn process count guard
• Hardened close() with try-finally + Unix pkill -P fallback
• Count-based orphan reaper in ProcessManager
• Circuit breaker (3 failures → 60s cooldown)
• etime-based sorting for process guards

Files restored to v10.0.2

• src/services/sync/ChromaSync.ts
• src/services/infrastructure/GracefulShutdown.ts
• src/services/infrastructure/ProcessManager.ts
• src/services/worker-service.ts
• src/services/worker/ProcessRegistry.ts
• tests/infrastructure/process-manager.test.ts
• tests/integration/chroma-vector-sync.test.ts

v10.0.3

Fix: Prevent chroma-mcp spawn storm (PR #1065)

Fixes a critical bug where killing the worker daemon during active sessions caused 641 chroma-mcp Python processes to spawn in ~5 minutes, consuming 75%+ CPU and ~64GB virtual memory.

Root Cause

ChromaSync.ensureConnection() had no connection mutex. Concurrent fire-and-forget syncObservation() calls from multiple sessions raced through the check-then-act guard, each spawning a chroma-mcp subprocess via StdioClientTransport. Error-driven reconnection created a positive feedback loop.

5-Layer Defense

Layer	Mechanism	Purpose
0	Connection mutex via promise memoization	Coalesces concurrent callers onto a single spawn attempt
1	Pre-spawn process count guard (execFileSync('ps'))	Kills excess chroma-mcp processes before spawning new ones
2	Hardened close() with try-finally + Unix pkill -P fallback	Guarantees state reset even on error, kills orphaned children
3	Count-based orphan reaper in ProcessManager	Kills by count (not age), catches spawn storms where all processes are young
4	Circuit breaker (3 failures → 60s cooldown)	Stops error-driven reconnection positive feedback loop

Additional Fix

• Process guards now use etime-based sorting instead of PID ordering for reliable age determination (PIDs wrap and don't guarantee ordering)

Testing

• 16 new tests for mutex, circuit breaker, close() hardening, and count guard
• All tests pass (947 pass, 3 skip)

Closes #1063, closes #695. Relates to #1010, #707.

Contributors: @rodboev

v10.0.2

Bug Fixes

• Prevent daemon silent death from SIGHUP + unhandled errors — Worker process could silently die when receiving SIGHUP signals or encountering unhandled errors, leaving hooks without a backend. Now properly handles these signals and prevents silent crashes.
• Hook resilience and worker lifecycle improvements — Comprehensive fixes for hook command error classification, addressing issues #957, #923, #984, #987, and #1042. Hooks now correctly distinguish between worker unavailability errors and other failures.
• Clarify TypeError order dependency in error classifier — Fixed error classification logic to properly handle TypeError ordering edge cases.

New Features

• Project-scoped statusline counter utility — Added statusline-counts.js for tracking observation counts per project in the Claude Code status line.

Internal

• Added test coverage for hook command error classification and process manager
• Worker service and MCP server lifecycle improvements
• Process manager enhancements for better cross-platform stability

Contributors

• @rodboev — Hook resilience and worker lifecycle fixes (PR #1056)

v10.0.1

What's Changed

OpenClaw Observation Feed

• Enabled SSE observation feed for OpenClaw agent sessions, allowing real-time streaming of observations to connected OpenClaw clients
• Fixed ObservationSSEPayload.project type to be nullable, preventing type errors when project context is unavailable
• Added EnvManager support for OpenClaw environment configuration

Build Artifacts

• Rebuilt worker service and MCP server with latest changes

v10.0.0 — OpenClaw Integration

OpenClaw Plugin — Persistent Memory for OpenClaw Agents

Claude-mem now has an official OpenClaw plugin, bringing persistent memory to agents running on the OpenClaw gateway. This is a major milestone — claude-mem's memory system is no longer limited to Claude Code sessions.

What It Does

The plugin bridges claude-mem's observation pipeline with OpenClaw's embedded runner (pi-embedded), which calls the Anthropic API directly without spawning a claude process. Three core capabilities:

1. Observation Recording — Captures every tool call from OpenClaw agents and sends it to the claude-mem worker for AI-powered compression and storage
2. MEMORY.md Live Sync — Writes a continuously-updated memory timeline to each agent's workspace, so agents start every session with full context from previous work
3. Observation Feed — Streams new observations to messaging channels (Telegram, Discord, Slack, Signal, WhatsApp, LINE) in real-time via SSE

Quick Start

Add claude-mem to your OpenClaw gateway config:

{
  "plugins": {
    "claude-mem": {
      "enabled": true,
      "config": {
        "project": "my-project",
        "syncMemoryFile": true,
        "observationFeed": {
          "enabled": true,
          "channel": "telegram",
          "to": "your-chat-id"
        }
      }
    }
  }
}

The claude-mem worker service must be running on the same machine (localhost:37777).

Commands

• /claude-mem-status — Worker health check, active sessions, feed connection state
• /claude-mem-feed — Show/toggle observation feed status
• /claude-mem-feed on|off — Enable/disable feed

How the Event Lifecycle Works

OpenClaw Gateway
  ├── session_start ──────────→ Init claude-mem session
  ├── before_agent_start ─────→ Sync MEMORY.md + track workspace
  ├── tool_result_persist ────→ Record observation + re-sync MEMORY.md
  ├── agent_end ──────────────→ Summarize + complete session
  ├── session_end ────────────→ Clean up session tracking
  └── gateway_start ──────────→ Reset all tracking

All observation recording and MEMORY.md syncs are fire-and-forget — they never block the agent.

📖 Full documentation: OpenClaw Integration Guide

---

Windows Platform Improvements

• ProcessManager: Migrated daemon spawning from deprecated WMIC to PowerShell Start-Process with -WindowStyle Hidden
• ChromaSync: Re-enabled vector search on Windows (was previously disabled entirely)
• Worker Service: Added unified DB-ready gate middleware — all DB-dependent endpoints now wait for initialization instead of returning "Database not initialized" errors
• EnvManager: Switched from fragile allowlist to simple blocklist for subprocess env vars (only strips ANTHROPIC_API_KEY per Issue #733)

Session Management Fixes

• Fixed unbounded session tracking map growth — maps are now cleaned up on session_end
• Session init moved to session_start and after_compaction hooks for correct lifecycle handling

SSE Fixes

• Fixed stream URL consistency across the codebase
• Fixed multi-line SSE data frame parsing (concatenates data: lines per SSE spec)

Issue Triage

Closed 37+ duplicate/stale/invalid issues across multiple triage phases, significantly cleaning up the issue tracker.

v9.1.1

Critical Bug Fix: Worker Initialization Failure

v9.1.0 was unable to initialize its database on existing installations. This patch fixes the root cause and several related issues.

Bug Fixes

• Fix FOREIGN KEY constraint failure during migration — The addOnUpdateCascadeToForeignKeys migration (schema v21) crashed when orphaned observations existed (observations whose memory_session_id has no matching row in sdk_sessions). Fixed by disabling FK checks (PRAGMA foreign_keys = OFF) during table recreation, following SQLite's recommended migration pattern.

• Remove hardcoded CHECK constraints on observation type column — Multiple locations enforced CHECK(type IN ('decision', 'bugfix', ...)) but the mode system (v8.0.0+) allows custom observation types, causing constraint violations. Removed all 5 occurrences across SessionStore.ts, migrations.ts, and migrations/runner.ts.

• Fix Express middleware ordering for initialization guard — The /api/* guard middleware that waits for DB initialization was registered AFTER routes, so Express matched routes before the guard. Moved guard middleware registration BEFORE route registrations. Added dedicated early handler for /api/context/inject to fail-open during init.

New

• Restored mem-search skill — Recreated plugin/skills/mem-search/SKILL.md with the 3-layer workflow (search → timeline → batch fetch) updated for the current MCP tool set.