topoteretes/cognee

This page introduces an external open-source repository. It is not an HDATF product.

What it does

Cognee is an open-source memory platform for AI agents. It ingests data in various formats, builds a self-hosted knowledge graph, and lets agents recall past context, decisions and learned rules across sessions.

How it helps ATF

Its way of linking a decision to the discussion and code behind it is worth comparing with Company Brain. Its agent memory across sessions could also be a reference for Harness.

License

Apache-2.0 Permissive, with a patent grant. Commercial use is allowed; keep the notices and state your changes.

More in this category

• TencentCloud/TencentDB-Agent-Memory
  TencentDB Agent Memory is a team memory hub for AI agents. It turns conversations, documents and code into Chat Memory, Skills, an LLM-Wiki and a Code-Graph that can be governed and shared across agents and frameworks.
• supermemoryai/supermemory
  Supermemory is a memory and context layer for AI that extracts facts from conversations, keeps user profiles, handles updates and contradictions, forgets expired information, and combines memory with document search. It can run fully locally.
• akitaonrails/ai-memory
  ai-memory is a self-hosted memory server shared across coding agents, so a task started in Claude Code can continue in Codex. Memory is kept as a git-backed Markdown wiki, with team sharing and an audit log.
• rohitg00/agentmemory
  A persistent memory server for coding agents such as Claude Code, Codex CLI, Cursor and Gemini CLI, and for any MCP client. It extends Karpathy's LLM Wiki pattern with confidence scoring, lifecycle, knowledge graphs and hybrid search.
• AgriciDaniel/claude-obsidian
  claude-obsidian is a local-first knowledge system for Claude Code that turns source material into linked, source-cited Obsidian pages in plain Markdown. It answers from evidence already in the vault and tracks each claim's support, contradictions and review state.

Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.