Get This Tool
Local RAG memory system
Pricing
- Model
- Free
Summary
Every session starts blank — your AI assistant has no memory of what you told it last week, what your preferences are, or what context you spent twenty minutes establishing. local-memory-mcp is a self-hosted MCP server that gives Claude and ChatGPT a persistent vector memory layer running entirely on your machine.
The server stores, retrieves, and versions memories using local ChromaDB, so context survives across sessions without touching any cloud service. You run it via Docker or Python, wire it into your MCP client once, and your assistant can recall preferences, project context, or past decisions on demand. Conflict detection flags when an incoming memory update collides with something already stored, so you are not silently overwriting context. The architecture fits solo developers and privacy-focused workflows well — it was built for exactly that. Where it strains: teams expecting multi-user memory sharing or production-grade scaling will find ChromaDB's local single-process model is not the right foundation.
Bottom line: Pick this if you want a zero-cost, local-only persistent memory layer for your personal Claude or ChatGPT workflow — hit its ceiling the moment you need shared memory across multiple users or a hosted API your team can all write to.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Fully local ChromaDB vector store with no external API calls, so your conversation history, preferences, and project context never leave your machine — a hard requirement for anyone working under data-residency or confidentiality constraints.
- MIT license with self-hosted Docker or Python install, which means zero ongoing cost and no vendor dependency — you are not one pricing change away from losing your memory layer.
- Built-in conflict detection when new memories contradict stored ones, so weeks of accumulated context does not get silently corrupted by a contradictory update.
- Stdio and HTTP/SSE transport options ship out of the box, so you can wire it into Claude Desktop as a local subprocess or run it as a persistent server depending on your workflow.
- Version tracking on stored memories, so you can audit what your assistant knows and roll back context that has gone stale — something absent in session-only assistants where there is nothing to audit at all.
Cons
Sign in to edit- ChromaDB runs as a local single-process store, which means the first time two MCP clients try to write memories concurrently — say, Claude Desktop and a script running in parallel — you hit locking contention. Teams building any multi-client or multi-user setup will need to replace ChromaDB with a server-backed vector store, at which point they are maintaining a fork.
- The docs describe no authentication or access control on the MCP server endpoint. Running this on anything other than localhost exposes the memory store to anyone on the same network. Adding auth is a code change, not a config toggle — teams with shared environments will build that themselves or choose a memory server that ships with it.
- Community activity is minimal at the time of curation — five stars, zero open issues, zero pull requests, seventeen commits. If a ChromaDB version bump breaks compatibility or an MCP spec update requires a transport change, there is no active maintainer cadence documented. Teams who need a maintained dependency in a production context will move to a more actively developed project.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- Docker, Python
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- 2026-06-18T03:37:43.471Z
Best For
Who it's for
- Users wanting local-only memory for MCP clients
- Privacy-focused AI assistant workflows
- Developers building persistent context layers
What it does well
- Maintain project context across Claude or ChatGPT sessions
- Store user preferences and schedules persistently
- Version and update memories with conflict detection
- Self-host RAG memory without cloud services
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Local RAG memory system
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is Local RAG memory system free?
- Yes — Local RAG memory system is fully free to use. There is no paid tier.
- Is Local RAG memory system open source?
- Yes. Local RAG memory system is open source.
- Does Local RAG memory system have an API?
- Yes. Local RAG memory system exposes a developer API. See the official documentation at https://github.com/ptobey/local-memory-mcp for details.
- Can I self-host Local RAG memory system?
- Yes. Local RAG memory system supports self-hosting on your own infrastructure.
- What platforms does Local RAG memory system support?
- Local RAG memory system is available on: Docker, Python.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
AI assistants lose all context when a session ends — preferences, project history, standing instructions, everything reset. local-memory-mcp solves this by running a local MCP server backed by ChromaDB vector search. You tell your assistant something once; the server stores it as a vector embedding on disk. On the next session, the assistant queries the server and retrieves relevant context before responding. The setup runs over stdio or HTTP/SSE transport, and both entrypoints ship as ready-to-run Python scripts with a Docker Compose file for containerized installs.
The differentiating decision here is that nothing leaves your machine. There is no hosted API, no account registration, no cloud vector store. The MIT license means you can fork it, modify it, and embed it in other projects without restriction. Version tracking and conflict detection are built into the memory layer — when a new memory contradicts an existing one, the server surfaces that collision rather than silently overwriting, which matters if your assistant is accumulating preferences over weeks.
This fits one profile precisely: a solo developer or privacy-conscious individual who wants persistent context for personal assistant workflows and does not want a subscription or external data dependency. It does not fit teams who need a shared memory store that multiple clients write to concurrently, or production deployments where ChromaDB’s local process model would become a bottleneck. The docs describe no authentication layer on the MCP server, so exposing it beyond localhost requires you to add that yourself.
The repository ships with Docker, a Dockerfile, a docker-compose.yml, and a .env.example, so the path from clone to running server is documented. MCP compatibility covers Claude Desktop and ChatGPT with MCP support; any other MCP-compatible client should work against the same transport layer.