$ loading_
Diagnose RAG pipeline regressions by pinpointing failure modes with statistical rigor.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "RAG Regression Gate" yet — see the docs or source repo.
Analyze the differences between this RAG evaluation run and the baseline. Identify which failure modes regressed significantly, especially retrieval miss, context mismatch, and hallucination, and use statistical methods to show whether the changes are significant.
Returns regression findings for each failure mode, significance judgments, impact levels, and prioritized root-cause investigation directions.
Compare the current candidate version with the production baseline. If only the overall score drops without significant worsening in failure modes, mark it as low risk; if a certain error type increases significantly, explain why the release should be blocked.
Provides a pass/fail regression gate decision with risk level, blocking reasons, and an evidence summary.
We updated the retrieval index and system prompt. Compare RAG results before and after the change, determine whether issues mainly come from retrieval or generation, and identify any newly significant regression patterns.
Outputs stage-by-stage regression diagnostics, showing affected failure modes and evidence linking them to index or prompt changes.
Index documents and retrieve relevant context for better LLM responses.
Let AI securely query private local documents with persistent memory.
Use authenticated MCP tools for graph-augmented and hybrid RAG retrieval.
Intelligent RAG tool that chooses between private knowledge and web search.
Retrieve relevant context and metadata from Qdrant using natural language queries.
Aggregate MCP servers and find tools through natural language semantic search.