mem0
66k stars · Open source · Python
The most-starred memory layer in the index, used to add persistent memory across sessions and users to agents you are building.
— You only need simple conversation history in a single session.
The shortlist
Give your AI agent long-term memory
Most agents forget everything when a session ends. A memory layer stores what the agent learns and brings the relevant parts back later. The options range from a library you build into your own agent to plain files your coding assistant can read.
Basic Memory
4.0k stars · Open source · Python
Stores what the assistant learns as local Markdown notes linked into a knowledge graph, which you can open and edit yourself, and works with Claude, Codex, Cursor and ChatGPT.
— Your organisation can't adopt AGPL-3.0 software.
Engram
6.8k stars · Open source · Go
Project memory for coding agents in a single Go binary with SQLite full-text search, exposed over MCP, with no Node.js, Python or Docker to install.
Before you commit
Memory is where agents quietly go wrong: a mistaken fact saved once gets retrieved and repeated for months. Prefer stores you can inspect and correct, keep personal or client data out unless you control where it is kept, and prune what the agent has saved from time to time.
More decisions
- Automate browser tasks
- Connect your coding agent to GitHub
- Research the web with citations
- Build a production AI agent
- Build a customer-support agent
- Build a voice receptionist
- Run an agent locally
- Automate work without code
- Feed your agent clean web data
- Give your agent live docs & search
- Pick a coding agent
- Build a knowledge assistant over your docs
- Orchestrate a team of agents
- Connect an AI agent to your database
- Turn Figma designs into code
- Use an AI agent inside your game engine