Open sourceTypeScript · ≈119k stars · Apache-2.0
claude-mem
Persistent context across sessions for every agent.
claude-mem runs a second agent beside your coding agent that records what it does and compresses it into observations; later sessions start with a short index and pull in details only when needed. It records sessions for Claude Code, Codex, Cursor, OpenCode, Antigravity, Pi, DeepSeek Harness, OpenClaw and others, while Copilot CLI, Goose, Roo Code, Warp and Zed get memory search only. It also includes My Work, a viewer of your requests and their status.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- DocsAutomatically captures tool-use observations, generates semantic summaries and injects them into future sessions.[1]
- DocsSupports many hosts: Claude Code, OpenCode (including 2), T3 Code, Antigravity CLI, Pi, DeepSeek Harness, OpenClaw and Grok Bot.[1]
- DocsLets you pick the memory provider after sign-in: the hosted observer, your own OpenRouter key, or your existing plan.[1]
- MetadataApache-2.0, about 104k stars and very frequent releases (v15.5.0 on 2026-10-29).[1]
Cons
- DocsThe installer asks you to sign in to a hosted 'observer' with a 30-day trial, and for Grok Bot the default is hosted CMEM Pro; the local observer is opt-in.[1]
- User reportsIt runs a separate worker service that must start for hooks to work, and Windows users hit worker start failures on port 37777 (2025, closed).[1][2]
- User reportsRelease regressions: v13.24.0 shipped the previous bundle and killed the worker on every hook event (September 2026), and 6.0.5 broke the plugin with a missed DB migration (2025).[1][2]
- Independent testA cold-start benchmark by the author of a rival tool found it does nothing out of the box on existing history and only records from install onward.[1]
- Best for
- Claude Code users who want automatic cross-session memory with minimal manual note-taking.
- Watch out for
- Check whether you are on the local or hosted provider, and pin versions given the history of breaking releases.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
People
Creator of claude-mem
Claude Code · Skills, plugins & mods · Context engineering & memory
Activity
- Last push
- · yesterday
- Latest release
- v16.0.2
· yesterday - Last 90 days
- 20+ releases
- · yesterday
Critical fix An auto-update no longer deletes the plugin version a running Claude Code session was loaded from (#6352, thanks @0xABCD01, closes #6349 reported by @justindrp). The cache sweep keeps a version folder while the Claude Code process recorded in its .in_use file is…
- · 2d ago
Critical fixes Signing in to Codex works again on Windows while claude-mem runs (#6332, thanks @ffe111, closes #6331). The private Codex home that claude-mem's app-server uses got its auth.json through a hard link to your real ~/.codex/auth.json. Codex refuses to replace a…
- · 3d ago
My Work puts what needs you first My Work, the page that shows what your agents are doing, was rebuilt from a full UI review (#6210, #6304). What needs you sits at the top, long lists fold, and every page works on a phone, by keyboard and with a screen reader. What needs you…
- · 4d ago
claude-mem no longer turns off Claude Code's Code Mode Code Mode starts with claude-mem enabled (#6278, thanks @naive000; reported in #6209 by @nasif-naseef). Code Mode (experimental, Claude Code 2.1.323 and later) refuses to start when a plugin's hooks module hooks tool.call.…
- · 4d ago
Two critical fixes The worker starts when settings.json is not valid JSON (#6259). The worker refused to start when ~/.claude-mem/settings.json couldn't be parsed, even though the database already held the device identity it was trying to save there. Nothing was captured and no…
Checked against GitHub on Nov 21, 2026. Release notes are shortened to plain text.
In research (1)
- Stratum: Repository Memory for Coding Agents Beyond the Context WindowVerdict: Incremental
Oct 26, 2026 · Shipped persistent session memory for Claude Code, and a direct practical alternative to Stratum's experiential tier.
Neighbours on the workbench
JavaScript · ≈308k stars · MIT
affaan-m/ECC
Agent harness performance optimization system: skills, instincts, memory and security for coding agents.
Maintained by Affaan Mustafa
TypeScript · ≈265k stars · MIT
NousResearch/hermes-agent
The agent that grows with you.
Maintained by Nous Research
v0.23.3releasedPython · ≈151k stars · Apache-2.0
Graphify-Labs/graphify
A /graphify skill that turns a codebase, docs and schemas into a queryable knowledge graph.
Maintained by Graphify-Labs
v0.9.114releasedConnections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Maintained by · 1
- thedotmackOrganisation
Built on · 3
- Claude CodeProject
- CodexProject
- OpenClawProject
Similar entries · 4
- agentmemoryProject
- Memorax CodeProject
- OpenCode SupermemoryProject
- opencode-memProject
Sources
Verified Nov 16, 2026
