MCP.so
Sign In
L

Linksee Memory

@michielinksee

Repository

About Linksee Memory

Local-first agent memory MCP — Claude Code, Cursor, OpenAI Codex, and Gemini CLI share one SQLite brain. 6-layer structured memory (goal/context/emotion/implementation/caveat/learning), drift detection with pre-action re-injection guard, and an AST-aware diff cache saving 50-99%

#memory#cross-agent#local-first#drift-detection#sqlite#claude-code#cursor#gemini-cli

Basic information

Category

Memory & Knowledge

Transports

stdio

Publisher

michielinksee

Submitted by

Michie Yamaguchi

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "linksee-memory": {
      "command": "npx",
      "args": [
        "-y",
        "linksee-memory"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Linksee Memory?

Linksee Memory is a local-first, cross-LLM memory MCP server that stores structured memory in a single SQLite file. It is designed to prevent AI agents from forgetting session context and drifting from past decisions, and works with Claude Code, Cursor, Windsurf, OpenAI Codex, and Gemini CLI.

How to use Linksee Memory?

Install the server with npx linksee-memory-setup. No further configuration or invocation commands are detailed in the README.

Key features of Linksee Memory

  • 6-layer structured memory: goal / context / emotion / implementation / caveat / learning
  • Drift detection (Intent Datadog) flags unrecorded divergences from declared decisions
  • Re-injection guard re-surfaces locked decisions before the agent acts
  • read_smart AST-aware diff cache saves up to 86% tokens on re-reads
  • Full MCP surface: Tools + Resources + Prompts + Sampling + Roots + Elicitation
  • Pain records (caveats) are never auto-forgotten

Use cases of Linksee Memory

  • Prevent agent forgetfulness across sessions by persisting structured memory
  • Detect and alert on decision drift when reality diverges from declared anchors
  • Share memory across multiple LLM tools without losing context
  • Reduce token consumption on repetitive re-reads via intelligent diff caching

FAQ from Linksee Memory

What problem does Linksee Memory solve?

It solves agent forgetfulness and silent drift across sessions by providing persistent, structured memory stored in a local SQLite file.

Which LLM agents are supported?

Claude Code, Cursor, Windsurf, OpenAI Codex, and Gemini CLI — all read from the same SQLite file.

How does drift detection work?

It allows users to declare decisions as anchors; when reality diverges without a recorded resolution, the server flags the discrepancy.

Is this a remote service or local-first?

Local-first. All memory is stored in one SQLite file on the user's machine.

What are the runtime dependencies?

A node environment is required to run the setup command; the server itself uses SQLite (no external database server).

Comments

More Memory & Knowledge MCP servers

Jupyter Notebook MCP Server (for Cursor)

jbeno

Model Context Protocol (MCP) server designed to allow AI agents within Cursor to interact with Jupyter Notebook (.ipynb) files

157

minutes

silverstein

Every meeting, every idea, every voice note — searchable by your AI. Open-source, privacy-first conversation memory layer.

1.3K

JupyterMCP - Jupyter Notebook Model Context Protocol Integration

jjsantos01

A Model Context Protocol (MCP) for Jupyter Notebook

130

MemoryMesh

CheMiguel23

A knowledge graph server that uses the Model Context Protocol (MCP) to provide structured memory persistence for AI models.

348

mcp-local-rag

nkapila6

"primitive" RAG-like web search model context protocol (MCP) server that runs locally. ✨ no APIs ✨

128