Wire transcript-derived coding-agent memory with deja-vu
Use deja-vu when Codex, Claude Code, Cursor, OpenClaw, Copilot, and other local coding agents need shared recall from session history already written on disk.
npx skills add agentskillexchange/skills --skill wire-transcript-derived-coding-agent-memory-with-deja-vu
Use deja-vu to index existing local coding-agent transcripts, redact sensitive material, wire recall into supported agents through MCP, plugins, skills, or CLI guidance, and answer future sessions with prior decisions, failed commands, file context, and handoff evidence. Invoke this when a later agent should automatically remember fixes, tradeoffs, and commands from previous local sessions without the operator manually searching each product’s history. The scope boundary is transcript-derived local recall and cross-agent memory wiring; it is not a generic vector database, hosted memory service, or note-taking product.
What this skill actually does
Inputs and prerequisites: deja-vu CLI, local coding-agent transcript stores, optional MCP-capable agents.
Setup notes: Install with Homebrew, Go, npm (npx @vshulcz/deja-vu), Scoop on Windows, a release bundle, or the upstream install script. Run deja install –auto when the operator wants deja-vu to wire supported agents automatically.
Source and verification boundary: use https://vshulcz.github.io/deja-vu/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.