memory_kg

Expert knowledge for installing, configuring, and using MemoryKG — a hybrid semantic + structural knowledge graph for document corpora (.md and .txt files). Use this skill when the user asks about: setting up MemoryKG in a project, adding memory-kg as a Poetry dependency, building the SQLite or LanceDB knowledge graph from documents, running the multipass analysis pipeline (memorykg pipeline run/embed/manifold), configuring .mcp.json for Claude Code or Kilo Code, configuring .vscode/mcp.json for GitHub Copilot, configuring claude_desktop_config.json for Claude Desktop, using the memorykg CLI (memorykg build, memorykg build-graph, memorykg build-index, memorykg query, memorykg pack, memorykg analyze, memorykg semantic-analyze, memorykg pipeline, memorykg viz, memorykg mcp, memorykg snapshot), using the graph_stats / query_docs / pack_docs / get_node MCP tools, or troubleshooting MemoryKG errors.

Expert knowledge for KGRAG — the unified cross-KG registry and federated query layer for CodeKG, MemoryKG, and MetaKG. Use this skill when: (1) Setting up or configuring KGRAG in projects, (2) Querying across multiple knowledge graphs simultaneously, (3) Extracting code/doc snippets for LLM context, (4) Integrating with Claude Code or Claude Desktop, (5) Running architectural analyses, (6) Managing the KG registry, or (7) Troubleshooting KG-related issues.

Use KGRAG to query across federated knowledge graphs (code, docs, metabolic) and manage cross-KG registries. Triggers when: searching multiple repos/KGs simultaneously, initializing KGs in new repos, querying code patterns across projects, extracting multi-KG snippets, or managing registry status.