📡 MCP Radar

tinhien11/tdai-memory-mcp

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.

⭐ 4 stars
⑂ 0 forks
JavaScript
momentum ▲ 8.0
created 2026-08-10
on radar since 2026-08-12
agent-memoryai-agentai-memoryclaude-codecodegraphcodexcursordeveloper-toolsdevinknowledge-graphlocal-firstlong-term-memorymcpmodel-context-protocolpersistent-memorysqlitetree-sittertypescript
View on GitHub ↗Homepage ↗

About tdai-memory-mcp

The only memory tool that understands your code. Not just text — symbols, callers, impact, errors, and the decisions behind them.

3 sessions, 3 React bugs, 1 memory: Session 1 captures the fix → Session 2 starts with it loaded → Session 3 knows both fixes.

Web viewer: Memory + CodeGraph + Wiki in one UI

Every memory tool stores text. Mem0, Zep, Letta, PMB, Claude Code's MEMORY.md — they all treat code as words. None of them know what a function is, who calls it, or what breaks when it changes.

tdai-memory-mcp is different. It has three layers in one SQLite file:

Layer What it knows Example --------- Memory Decisions, bugs, learnings, errors "We chose SQLite over Postgres for local-first storage" CodeGraph Symbols, callers, callees, impact useEffect() is called by 47 components, calls cleanup() Wiki Markdown docs, ADRs, outdated detection "ADR-007 says use SQLite. 3 files still import Postgres."

One recall("useEffect") returns the bug fix you did last week + the function that caused it + who calls it + the ADR that decided the pattern. No other tool does this.

That is it. Auto-detects Claude Code, Devin, Cursor, Codex. Registers MCP server + hooks. Restart your agent.

Global install (faster hooks, no npx delay): npm install -g tdai-memory-mcp — postinstall auto-runs setup.

Bug fix chain — 3 sessions, 3 React bugs, 1 memory

A React project had 3 bugs in the fiber reconciler. Each session fixed one. The next session started with the previous fix already loaded.

From the project README.

Maintaining this server?

Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:

[![On MCP Radar](https://mcp.liqiwa.com/badge.svg)](https://mcp.liqiwa.com/s/tinhien11--tdai-memory-mcp.html)

Related MCP servers

tinhien11/remem-mcp

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.

⭐ 3TypeScript
friday-memory/friday

⚡ The open-source persistent cognitive memory layer for AI coding agents. Stop your Cursor, Claude & Copilot from having amnesia.

⭐ 51HTML
Haroon-jay/wctx

Cross-repository context for coding agents. A local-first MCP server that turns finished sessions into findings with commit provenance and staleness checks.

⭐ 7TypeScript
Kerneta/daidocs

Open 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

⭐ 29JavaScript
Koragraph/KoragraphMCP

The memory layer that gets better as you use it. What broke, why, what depended on what, even the "push this to Wednesday" kind of thing that never makes it into any doc. No one has to write it down.

⭐ 13JavaScript
Masikomoore/agent-memory

One memory for every AI agent — self-hosted long-term memory over MCP/REST for Claude Code, Codex, Cursor, Gemini CLI, VS Code, OpenClaw, and more.

⭐ 3JavaScript

📬 Get the weekly radar in your inbox

The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.