$ loading_
Trace multi-graph causal paths to answer why questions with confidence scores.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "polyg-mcp" yet — see the docs or source repo.
Using the multi-graph memory, analyze why this agent recommended option B instead of option A in the last three conversation turns. Trace key evidence across semantic, entity, temporal, and causal relations, and return confidence-scored paths.
Returns the key causal chains behind recommending option B, relevant memory nodes, and confidence scores for each path.
Based on stored events and entity relations, answer: why did the conversion rate drop this week? Connect the timeline, related business events, and potential causal factors, then rank the results by confidence.
Outputs multiple explanation paths for the conversion drop, including timeline order, contributing factors, and confidence ranking.
Analyze why this defect appeared after release and was not caught by testing. Combine commit history, requirement changes, test coverage, and historical issues to generate cross-graph root-cause paths.
Provides causal explanation chains for defect introduction and missed detection, with key nodes and confidence scores.
Provides persistent graph memory for LLMs with auto-linking and layered recall.
Build and query persistent knowledge graphs so coding agents remember across sessions.
Persistent knowledge-graph memory for MCP with semantic search and version tracking.
Build a semantic graph from project files for search, knowledge, and task management.
Provide persistent graph-memory storage and retrieval for LLM agents via MCP.
Give AI coding assistants memory, code graph insight, and safe multi-agent coordination.