RLM Memory MCP
TypeScript · MCP
Overview
A Model Context Protocol server giving AI agents persistent memory and semantic file discovery, long-term recall for agentic workflows.
RLM Memory MCP is an open-source Model Context Protocol server that gives AI agents persistent memory and semantic file discovery. It acts as a dedicated memory layer that agents can read from and write to across sessions, so context is never lost when a workflow restarts.
The server is built entirely in TypeScript and exposes a clean MCP interface. Agents query it to retrieve semantically relevant memories and locate files by meaning rather than exact path, which removes a common bottleneck in long-running agentic pipelines.
Persistent memory is one of the most practical gaps in current agentic tooling. RLM Memory MCP addresses that directly, making it possible to build agents that accumulate knowledge over time and behave consistently without needing every fact re-injected at each invocation.
- Type
- TypeScript · MCP
- Year
- 2025
- Status
- Open Source
- Sheet
- A-05
- MCP server
- Persistent memory
- Semantic file discovery
- Long-term recall
- TypeScript
- MCP
- Vector search
- Persistent memory across sessions
- Semantic file discovery
- Built in TypeScript
- Open source, Model Context Protocol
- Long-term recall for agentic workflows