mcp · handle cowork_semantic_search · source page
ken · job 74%sense · job 77%obsidian_graph · job 75%moss · job 76%alexandria · job 75%Knowledge Raven · job 78%directory_indexer · job 78%repobase · job 78%mcp_docs_provider · job 72%ai_search · job 71%neuroncite · job 78%codebase_rag · job 70%mcp_documentation_server · job 79%ragie · job 75%local_faiss_mcp · job 74%Toolbelt · job 81%sunbeam_memory · job 71%graph_memory · job 73%gortex · job 75%mnemodoc_server · job 79%searchstack · job 76%marm_memory · job 75%pluck · job 77%codesearch · job 74%code_context_engine · job 73%sourcerer_mcp · job 71%qdrant_loader · job 71%ontomics · job 71%mcp_local_rag_2 · job 72%lybrary · job 75%Local RAG MCP · job 80%code_memory · job 77%Local Research Library MCP · job 78%MCP RAG Server · job 77%DocSift · job 76%Marrine RAG MCP Server · job 75%Keepygaga RAG · job 79%omniclip_rag · job 76%SVDB MCP · job 71%SuperAIDB · job 73%Personal Knowledge-Base MCP Server · job 74%Hybrid RAG MCP · job 71%Corpus · job 72%OneDroid Engram · job 75%Percentages are graph-similarity metadata, not suitability scores. A graph edge answers “is it connected” — plan fit and comparable cost are checked below.
Pick the exact reference and alternative series. Each history is separate; gaps stay gaps.
This view calls the same get_evidence tool your agent would — chart, table and MCP response describe the same records. Watch the reference series