About Memoryvault Kit
Memoryvault Kit is an MCP server in the RAG category: a personal memory layer for your AI tools. Intelligent authoring (gap detection, self-enriching graph, session-synthesis loop) + measurably good retrieval (94.9% blind Cov@10, sub-ms latency). Domain-agnostic framework — works for any context you want to track. MCP-native, Claude-Code-first, works with Cursor/Continue/Cline/OpenAI/Gemini.
Install
git clone https://github.com/ayushmall/memoryvault-kitThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (7)
/memory-graph-auditWalks you through Obsidian's graph view to catch what code can't see
Parameters not published.
evals/retrieval/retrievers/*.pyRun a specific retriever variant against the 482-Q hardened set
Parameters not published.
memory doctor --eval-recovery5 structural checks before the eval runs
Parameters not published.
memory doctor --signal-qualityPer-source ingest-vs-retrieval noise ratio
Parameters not published.
memory evalThree-pillar score (fill_quality + pollution + consistency)
Parameters not published.
memory eval --softCoverage: % of questions returning ≥2 results scoring ≥5. No gold annotations required
Parameters not published.
memory eval init --from-vaultGenerate a starter eval set from your actual vault content
Parameters not published.
Ask AI
Ask AI about Memoryvault Kit
Powered by Claude · Grounded in docs
Security checks
6 checks have not been run yet.