topoteretestopoteretesOfficialยท10 Agent Skills Included

cognee

Give AI agents persistent long-term memory with knowledge graphs

Turns documents, files, URLs, and databases into a searchable knowledge graph that gives AI agents persistent memory across sessions. Replaces fragile RAG pipelines and manual context management with automatic entity extraction, graph reasoning, and feedback-driven self-improvement. Ships SDK, CLI, API server, and Docker images so agents remember, recall, and learn from every interaction.
npx skills add topoteretes/cognee --all -g -y
Available:

Tells the AI agent how cognee's memory API, pipeline architecture, database adapters, and multi-tenant modes work so it can operate and extend the platform correctly.

All Skills in This Repository (10)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install cognee?โ–ผ

Run `npx skills add topoteretes/cognee --all -g -y` in your terminal to install all skills in this suite globally.

What does cognee do for AI agents?โ–ผ

Cognee gives agents persistent long-term memory by turning your documents and data into a knowledge graph they can search and recall across sessions. It combines vector search, graph reasoning, and feedback loops so agents improve over time without retraining.

How is cognee different from traditional RAG?โ–ผ

Instead of retrieving flat text chunks, cognee builds a knowledge graph with extracted entities and relationships, enabling multi-hop reasoning and memory that evolves as your data grows.

Does cognee work with Claude Code and other AI tools?โ–ผ

Yes. Cognee ships an MCP server, a Claude Code plugin, and an OpenClaw plugin, so agents in Claude Code, Cursor, and similar tools can use its memory directly.

Can I run cognee locally without cloud services?โ–ผ

Yes. Cognee runs fully locally with embedded SQLite, LanceDB, and Ladybug databases by default, and Docker images are available for one-command setup.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’