Local, git-versioned memory for AI coding agents. No RAG, no Docker, no external service. Capture, compile, recall over a local LLM wiki with on-device embeddings and an MCP server.
Skill Claude CodeCodex
Debug llm-wiki-memory capture/recall — flush breadcrumbs, daily-vs-compile lag, failed-distill stash recovery, embedding cache, hook wiring. Use when memory didn't capture, doesn't recall, or a hook/cron looks broken.
Skill Claude CodeCodex
Run the llm-wiki-memory test suite without filling /tmp — sweep stale lwm- workspaces, iterate on single files, run npm test once. Use when running or debugging tests in this repo.
Skill Claude CodeCodex
Skill: triage review findings without acting on false positives — Full criteria for the guard-deletion case below: .agents/rules/defensive-invariants.md. Canonical file is .agents/skills/triage-review-findings.md.
Skill Claude CodeCodex
Skill: verify a change to the live engine tree — /.llm-wiki-memory/src runs LIVE — the MCP server, the Claude Code hook workers, and cron import it as you edit. Canonical file is .agents/skills/verify-live-change.md.
Skill Claude CodeCodex
Author an idempotent, state-detecting migration under scripts/migrations/ for a breaking llm-wiki-memory change. Use when shipping a breaking change to the engine, or when asked to write a migration / upgrade path. Replaces the retired release-runbook flow.