Who it is forA developer who wants to try graph memory without paying for or sending data to an LLM.
The problemMost memory layers need a hosted model and API key before anything works.
The outcomeA local dataset you can write to and query from the command line.
What you need first
- Python 3.10 to 3.14; install with `uv pip install "cognee[gliner]"` for the local extraction model
- Optional: an LLM key (LLM_API_KEY) for generated answers instead of retrieved passages
- Optional: Docker for the API, UI and MCP server; Node.js/npm for the local UI launcher
Steps
- Install with
uv pip install "cognee[gliner]" - Store a fact with
cognee-cli remember "Marie Curie was born in Warsaw." -d local_quickstart - Query it with
cognee-cli recall "Where was Marie Curie born?" -d local_quickstart - Or save the README quickstart as quickstart.py and run
python quickstart.py
Prompt for your agent
Using topoteretes/cognee, install `cognee[gliner]`, remember three facts from <my notes> into a dataset with cognee-cli, recall one of them, and show me the retrieved text.
Caveat. Without an LLM key you get retrieved passages, not generated answers; the local models download on first use.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for topoteretes/cognee and "Build agent memory locally with no API key"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer