MCPVault

retrieval_diagnostics

MCP tool from Mind Mem by star-ga

Pipeline rejection rates, intent histogram, hard negatives

How to use it

retrieval_diagnostics is exposed by the Mind Mem MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the retrieval_diagnostics tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.

Install Mind Mem

$pip install mind-mem
FULL MIND MEM LISTING

Other tools in Mind Mem (20)

approve_apply

Apply a staged proposal (dryrun=True by default)

calibration_feedback

Submit quality feedback for a retrieved block (thumbs up/down)

calibration_stats

View per-block and global calibration statistics

category_summary

Category summaries relevant to a given topic

delete_memory_item

Delete a memory block by ID (admin-scope)

export_memory

Export workspace as JSONL (user-scope)

find_similar

Find blocks similar to a given block

get_mind_kernel

Read a specific MIND kernel configuration as JSON

hybrid_search

Hybrid BM25+Vector search with RRF fusion

index_stats

Index statistics, MIND kernel availability, block counts

intent_classify

Classify query intent (9 types with parameter recommendations)

list_contradictions

List contradictions with auto-resolution analysis

list_mind_kernels

List available MIND kernel configurations

memory_evolution

View/trigger A-MEM metadata evolution for a block

prefetch

Pre-assemble context from recent conversation signals

propose_update

Propose a decision/task — writes to SIGNALS.md only

recall

Search memory with BM25 (query, limit, activeonly)

reindex

Rebuild FTS5 index (optionally including vectors)

rollback_proposal

Rollback an applied proposal by receipt timestamp

scan

Run integrity scan (contradictions, drift, signals)