deja-vu MCP Server
vshulcz/deja-vu
[](https://agentrank-ai.com/tool/vshulcz--deja-vu/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) Memory for Claude Code, Codex, Cursor and 38 more coding agents, built from the session history already on your disk. Local search, MCP and hooks, no LLM, one Go binary.
claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
vshulcz/deja-vu is a Go tool licensed under MIT. Memory for Claude Code, Codex, Cursor and 38 more coding agents, built from the session history already on your disk. Local search, MCP and hooks, no LLM, one Go binary. Topics: claude-code, codex, mcp, opencode, agent-memory, coding-agents, mcp-server, claude-code-hooks, claude-code-plugins, session-history, deepseek-harness, dsh-plugin, gemini-cli-extension, kimi-code, qwen-code, zed-extension, agents, ai, genai, memory.
Ranked #1248 out of 24846 indexed tools.
In the top 6% of all indexed tools.
Has 1,185 GitHub stars.
Has 37 contributors.
Actively maintained with commits in the last week.
Ecosystem
Score Breakdown
1,185 stars → well-known project
Last commit 1d ago → actively maintained
1625/1646 issues closed → responsive maintainer
37 contributors → active community
4 dependents → few downstream users
Weights: Freshness 20% · Issue Health 20% · Dependents 22% · Stars 10% · Contributors 8% · How we score →
Matched Queries
From the README
<p align="center">
<picture>
<source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/vshulcz/deja-vu/main/assets/logo-dark.svg">
</picture>
</p>
<p align="center"><b>The one memory your coding agents share, built from the history already on your disk.</b></p>
<p align="center">Your agent is about to re-debug something you fixed in March — in a different agent.
deja indexes the sessions Claude Code, Codex, Cursor and every other agent on this machine
already wrote to disk, and hands the right one back in whichever agent asks.</p>
<p align="center"></p>
<p align="center"><sub><em>Nobody searched anything — the agent called deja itself. Two genuine runs, a real model and a real tool call, against a synthetic corpus: nobody's history is published.</em></sub></p>
<p align="center"><b>deja starts full: the history 34 agents already wrote, searchable while it indexes, with no model and no capture step.</b></p>
<p align="center">And nobody has Read full README on GitHub → Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.