Team-memory – your team's shared brain, auto-built from Claude Code CLI or UI
team-memory is a tool that creates a shared, version-controlled knowledge base for teams, automatically built from interactions with Claude Code or manual entries via a web interface. It stores all information in a Git repository owned by the team, using Markdown files that support full version control and direct editing. The system runs locally, integrates with GitHub, and emphasizes privacy by binding only to localhost and avoiding external telemetry.
- ▪team-memory stores team knowledge in a Git repository as plain Markdown files, enabling version control and direct editing.
- ▪It integrates with Claude Code to automatically summarize and commit session outcomes into the knowledge base.
- ▪The tool runs locally, binds only to 127.0.0.1, and does not send data to external servers or collect telemetry.
- ▪Users can access a web UI to manually add, browse, or flag outdated entries in the knowledge base.
- ▪Installation supports macOS, Linux, and Windows, with optional service mode for continuous availability.
Opening excerpt (first ~120 words) tap to expand
team-memory Shared, version-controlled knowledge base your whole team builds together Locally run in your CLI or UI, backed up to your git knowledge repo automatically Most team knowledge lives in people's heads, scattered Slack threads, or stale wikis nobody trusts. team-memory is different: it lives in a Git repo your team already owns, every entry is a commit, and anyone can open the web UI and jot something down in seconds. Two ways to build it up: Web UI — lightweight note-taking interface at http://team-mem.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.