# mmr_search A execute tool on the Mcpg MCP server. Agent View of the PolicyLayer registry record for `mmr_search`. HTML page: https://policylayer.com/tools/io-github-devopam-mcpg/mmr-search ## Facts - Tool: `mmr_search` - Server: Mcpg (`pypi:mcpg`) — https://policylayer.com/tools/io-github-devopam-mcpg.md - Install: `npx -y pypi:mcpg` - Homepage: https://github.com/devopam/MCPg - Risk category: Execute (High risk) - Registry record: grade F, identity unverified - Server rate-limited: no - Parameters: 0 - Recommended policy verdict: Rate-limited ## Example call (MCP tools/call, JSON-RPC 2.0) ```json { "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "mmr_search", "arguments": {} } } ``` ## Why mmr_search is rated High MMR (Maximal Marginal Relevance) search is typically a read/query operation used in vector similarity search to retrieve diverse results. However, because the description is empty, there is uncertainty about whether it executes arbitrary queries or has side effects. From the tool's own definition: "Tool name 'mmr_search' suggests a search/read operation (Maximal Marginal Relevance search), but the description is empty and uninformative. Sibling tools include vector search, reranking, and query analysis tools on a PostgreSQL MCP server." ## Use case AI agents invoke mmr_search to trigger actions in Mcpg. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started. ## Recommended policy (PolicyLayer) Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Mcpg: ```json { "version": "1", "default": "deny", "tools": { "mmr_search": { "limits": [ { "counter": "mmr_search_rate", "window": "minute", "max": 10, "scope": "grant" } ] } } } ``` ## Other tools on Mcpg (251) - `cancel_migration` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/cancel-migration.md - `drop_graph` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/drop-graph.md - `drop_property_graph` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/drop-property-graph.md - `drop_publication` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/drop-publication.md - `drop_subscription` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/drop-subscription.md - `dump_database` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/dump-database.md - `partman_drop_partition` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/partman-drop-partition.md - `prune_audit_events` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/prune-audit-events.md - `restore_database` — Destructive — https://policylayer.com/tools/io-github-devopam-mcpg/restore-database.md - `cancel_query` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/cancel-query.md - `cluster_vectors` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/cluster-vectors.md - `complete_migration` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/complete-migration.md - `dry_run_ddl` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/dry-run-ddl.md - `enable_extension` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/enable-extension.md - `enable_logical_replication_on_demand` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/enable-logical-replication-on-demand.md - `enable_redis_fdw` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/enable-redis-fdw.md - `generate_graph_projection` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/generate-graph-projection.md - `generate_test_data` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/generate-test-data.md - `hybrid_search` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/hybrid-search.md - `maintain_turboquant_index` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/maintain-turboquant-index.md - `open_cursor` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/open-cursor.md - `optimize_query` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/optimize-query.md - `partman_run_maintenance` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/partman-run-maintenance.md - `pg_search_run` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/pg-search-run.md - `prepare_migration` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/prepare-migration.md - `prewarm_recommended` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/prewarm-recommended.md - `prewarm_relation` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/prewarm-relation.md - `recommend_redistribute` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/recommend-redistribute.md - `reindex_pg_search_index` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/reindex-pg-search-index.md - `reindex_turboquant_index` — Execute — https://policylayer.com/tools/io-github-devopam-mcpg/reindex-turboquant-index.md - …and 221 more: https://policylayer.com/tools/io-github-devopam-mcpg.md ## For agents This record is a snapshot. Live verdicts and the full registry: - Check every server in your MCP config at once: `npx -y policylayer stack` - Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md) - Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events` --- Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=io-github-devopam-mcpg · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/io-github-devopam-mcpg