Team-memory – your team's shared brain, auto-built from Claude Code CLI or UI (github.com)

A new tool called team-memory has been introduced, designed to create a shared, version-controlled knowledge base that teams can collaboratively build upon. This solution addresses the common challenge of fragmented knowledge within organizations, often trapped in people's minds or scattered across ineffective communication channels like Slack or outdated wikis. Team-memory stores entries in a Git repository, with each note tracked as a commit, making it easy for team members to contribute via a simple web interface or through an automatic hook that captures session summaries when using Claude Code. This development is significant for the AI/ML community as it enhances collaborative efficiency and knowledge retention. Team-memory leverages Git-backed storage and smart categorization through large language models (LLMs) to facilitate easy note-taking, searching, and manual entry management. Its local operation, reliance on Markdown for simplicity, and commitment to privacy—binding only to localhost and avoiding telemetry—make it a secure alternative for teams seeking to maintain collective knowledge without sacrificing confidentiality. The tool's implementation is straightforward, requiring minimal setup and supporting both macOS and Linux, thus expanding its accessibility to a broad spectrum of users.