topoteretes/cognee
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine. observed · 2026-08-28
Health v2 · maintenance only
91/100
- Activity 99
- Release rhythm 87
- Longevity 79
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 4
- age_days: 1113
- days_rel: 10
- days_push: 7
- n_releases_24m: 104
Adoption not part of the score
30281 stars · 2969 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded
Cognee is an open-source Python library and platform that gives AI agents persistent long-term memory by ingesting data in any format and building a self-hosted knowledge graph combined with vector search. It ships SDKs, an HTTP API, MCP server, and plugins for tools like Claude Code, OpenAI Agents SDK, and Hermes, enabling agents to remember, recall, and reason across sessions.
Use cases
- give my ai agent persistent memory across sessions
- build a self-hosted knowledge graph from documents
- add long-term memory to claude code or coding agents
- implement graph rag for llm context enrichment
- store and recall agent decisions and past work
- build a company brain from docs, chats, and tickets
- connect an mcp memory server to my ai assistant
- index markdown notes and recall relevant context automatically
When to choose
- you need agents to remember facts, decisions, and context across sessions
- you want a self-hosted, open-source memory engine with graph + vector retrieval
- you use Claude Code, OpenAI Agents SDK, or MCP-compatible tools and want drop-in memory
- you need structured, ontology-aware memory rather than loose text embeddings
- you want a company knowledge base with citations and provenance
When to avoid
- you only need simple key-value session state with no semantic retrieval
- you want a fully managed hosted service with zero infrastructure
- your project is not Python or Node-based and you cannot run the API server
- you need lightweight in-process memory without a graph database dependency
Facets
library · maturity active
rag vector-database agent-framework mcp search-engine etl llm-inference artificial-intelligence large-language-models databases developer-tools python self-hosted cross-platform cli agent-memory knowledge-graph graph-rag memory-management context-engineering cognitive-architecture mcp-server long-term-memory self-hosted-ai semantic-layer ai-agents retrieval-augmented-generation knowledge-graphs docker
10 sources
- readme: https://github.com/topoteretes/cognee · fetched 2026-08-28 · ab619e9295d3
- homepage: https://www.cognee.ai · fetched 2026-08-29 · 90ac0b69aa57
- site_page: https://docs.cognee.ai · fetched 2026-08-29 · 7f9d0919ac17
- site_page: https://docs.cognee.ai/cognee-mcp/integrations/claude-code · fetched 2026-08-29 · 2249c1fa4b23
- site_page: https://docs.cognee.ai/integrations/openai-agents-sdk-integration · fetched 2026-08-29 · b1f93873c712
- site_page: https://docs.cognee.ai/getting-started/llm-quickstart-skill · fetched 2026-08-29 · ac1c2d69bbd0
- site_page: https://docs.cognee.ai/integrations/openclaw-integration · fetched 2026-08-29 · c905c6e6387f
- site_page: https://docs.cognee.ai/integrations/hermes-agent-integration · fetched 2026-08-29 · 3ff484c73783
- registry_pypi: https://pypi.org/pypi/cognee/json · fetched 2026-08-29 · 7a4ea68c6c46
- site_page: https://www.cognee.ai/pricing · fetched 2026-08-29 · 6ecfcdb68ec6
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| topoteretes/cognee | main | 91 |
For agents
markdown · JSON · MCP: product_card(name="topoteretes/cognee")
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem