$ loading_
Log and audit RAG retrieval in MCP agents for replay, diffing, and debugging.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "retrieval-lens" yet — see the docs or source repo.
Use retrieval-lens to log this MCP agent's RAG retrieval run and output the chunks seen, relevance scores, sources, and rankings at each step so I can find why the answer cited irrelevant content.
An auditable retrieval log listing candidate chunks, scores, sources, and rankings, plus likely causes of the retrieval issue.
Compare retrieval records for the same query under two different configurations, show differences in chunks, scores, sources, and rankings, and summarize which setup is more stable.
A retrieval diff report showing recall and ranking changes across both runs, with a clear configuration assessment.
Replay this historical RAG retrieval record in chronological order, reconstructing what the model saw, the retrieval sources, and ranking results for audit and review.
A chronological retrieval replay that helps the team audit exactly what information the model was exposed to.
Turn unstructured documents into a searchable knowledge base for AI agents.
Intelligent RAG tool that chooses between private knowledge and web search.
Search and add traceable RAG knowledge for each project workspace.
Use authenticated MCP tools for graph-augmented and hybrid RAG retrieval.
Let AI securely query private local documents with persistent memory.
Diagnose RAG pipeline regressions by pinpointing failure modes with statistical rigor.