Long-term memory MCP server for AI coding agents. SQLite + FTS5 + vector search, CodeGraph, error learning, compaction survival. Beats Mem0 on benchmarks. No cloud, no API key.
Your coding agent stops repeating the same mistakes.
Local memory that survives context compaction — learns from every error, injects fixes before the next attempt, and syncs to your git repo so your whole team shares it.
One command setup. No API key. No cloud. No database server. Just a SQLite file.
Error learning loop CodeGraph search Web viewer ---------
Viewer: overview CodeGraph: callers CodeGraph: search ---------
That's it. Auto-detects Claude Code, Cursor, Devin, Codex. Registers MCP server + hooks. Restart your agent.
The demo creates a real TypeScript project, runs real npm run build, captures real TS2307 errors, and shows the full learning loop — capture → inject → fix → zero retries. No hardcoded strings.
remem-mcp Mem0 Claude MEMORY.md Mneme --------------- Survives compaction Yes — PreCompact hook saves checkpoint, re-injects after Yes — cloud store No — 200-line cap, silent truncation Yes — PreCompact hook Learns from errors Yes — auto-captures, injects fixes No No No Semantic search Hybrid BM25 + sqlite-vec Vector only No — LLM filename picker, max 5 files Vector + graph Setup 1 command API key + cloud Built-in Build from source (Rust) Data location Local SQLite Cloud Local markdown Local SQLite Team sharing Git-native (commit, diff, merge) Cloud sync Copy-paste Manual API key No Yes No No Cost Free $19–249/mo Free Free
Then run npx remem-mcp install-hooks. MCP tools require sandboxmode = "danger-full-access".
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/tinhien11--remem-mcp.html)
Local-first MCP memory server — Memory + CodeGraph + Wiki in one SQLite file. Tree-sitter code indexing, markdown wiki, lifecycle hooks, auto-capture. No API key, no cloud.
Kerneta/daidocsOpen plain-text file format for AI memory. Your assistant's long-term memory as .dai files on your disk: readable by Claude, GPT, Gemini, Cursor, local models and grep (all LLM models work). MCP serve
Haroon-jay/wctxCross-repository context for coding agents. A local-first MCP server that turns finished sessions into findings with commit provenance and staleness checks.
osaxyz/sepiaceLong-term memory for coding agents. The sepiace plugin for Claude Code and Codex.
2672243194/agentbrainLocal-first long-term memory for AI agents - plain Markdown vault + thin MCP server. CJK-aware BM25, token-efficient by design.
abhishekswe/agent-fastpathJev MCP server: a decision layer for coding agents, built on TypeSafe Jev (System One model). Ship gates, risk checks, file triage that keeps files out of context, and a safe headless browser, with ca
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.