# db0

> The data layer for AI agents. Memory that evolves, state that recovers, context that stays current. Open-source SDK, local-first SQLite, no API keys.

## What db0 does

db0 gives AI agents durable memory that persists across sessions. It handles:

- **Memory**: Scoped storage (user, agent, session, task) with superseding so old facts retire when new ones arrive
- **State**: Checkpoints and branching so agents can recover from bad turns
- **Context**: Assembles the right knowledge into the token budget before each LLM call
- **Search**: Hybrid semantic + keyword search across stored memories
- **Consolidation**: Clusters related memories and merges them automatically — auditable, configurable, cheaper than auto-dream

## Key facts

- Open source, MIT licensed
- Written in TypeScript
- Storage: SQLite (local, default) or PostgreSQL (sync across devices)
- No signup, no API key, no cloud dependency
- Works offline
- Not a wrapper around an LLM — db0 does not call any model
- Alternative to mem0 for teams that need memory + state + context together

## Integrations

- **Vercel AI SDK**: Memory middleware that wraps any model. Install: `npm i @db0-ai/ai-sdk`
- **LangChain**: Drop-in memory tools and chat history. Install: `npm i @db0-ai/langchain`
- **OpenClaw**: Plugin that gives OpenClaw agents persistent memory that survives compaction. Install: `npx @db0-ai/openclaw init`
- **Claude Code**: 10 MCP tools for memory, state checkpoints, and semantic search. Install: `npx @db0-ai/claude-code init`
- **Pi**: Cross-session memory for Pi coding agent. Install: `npx @db0-ai/pi init`

## Resources

- What is AI agent memory: https://db0.ai/resources/what-is-ai-agent-memory
- RAG vs agent memory: https://db0.ai/resources/rag-vs-agent-memory
- Memory scopes for AI agents: https://db0.ai/resources/memory-scopes

## Links

- Website: https://db0.ai
- GitHub: https://github.com/db0-ai/db0
- npm: https://www.npmjs.com/org/db0-ai
- Blog: https://db0.ai/blog
- Compare to mem0: https://db0.ai/compare/mem0