Analyzes AI agent decision logs to find reasoning errors and suggests fixes to enhance performance and reduce hallucinations.
# Agent Reflect Engine A self-reflection engine that analyzes AI agent decision logs to identify reasoning flaws and suggest optimization patches, improving long-term performance and reducing hallucination. ## Usage ```bash # Analyze agent logs and get reflection report python agent_reflect_engine.py logs/agent_trace.jsonl --output report.json # Include trusted knowledge base for hallucination detection python agent_reflect_engine.py logs/agent_trace.jsonl --knowledge kb.json --output report.json # Pipe directly for automation python agent_reflect_engine.py logs/latest.jsonl | jq '.patch_suggestions' ``` ## Price $9.99
don't have the plugin yet? install it then click "run inline in claude" again.