Every memory tool starts empty and records forward. deja starts full. It indexes the sessions all 24 of your coding agents already wrote to disk — months of history from before you installed it — and serves them back over MCP, in whichever agent asks.

Twenty-five coding agents write every conversation to local files: Claude Code, Codex, Cursor, opencode, Gemini CLI, Cline, Copilot CLI, VS Code Copilot Chat, Roo Code, aider, Goose, Qwen Code, Kimi Code, Antigravity, Grok Build, OpenClaw, pi, omp, DeepSeek Harness, Hermes and Zed. deja turns those files into one memory layer that all of them can read.

One Go binary. No LLM, no embeddings, no API key, nothing leaves the machine. 85.3% hit@1 on LongMemEval-S, millisecond lookups over 5 GB of history.

npx @vshulcz/deja-vu "connection pool exhausted"   # search, no install
npm install -g @vshulcz/deja-vu                    # then: deja install --auto

deja install --auto wires MCP recall into every coding agent it finds on the machine and turns on session-start recall where the agent supports it. Ten seconds to install, about ten to index, and the next session already knows. Use --all for the MCP tools without the session-start hook.

Full documentation, the harness matrix and the benchmarks: github.com/vshulcz/deja-vu · vshulcz.github.io/deja-vu

MIT