{"slug":"agent-trace-triage","title":"agent-trace-triage","description":"Agent Trace Triage is a Claude Code plugin for developers building agentic AI systems (LangGraph, CrewAI, custom tool loops, MCP agents).\n\nWhen an agent run fails, stuck in a loop, wrong tool, or burning tokens. the user runs /agent-trace-triage with a JSON/JSONL trace file. The plugin:\n\n• Parses traces (JSONL, OpenAI tool_calls, partial LangSmith/OTEL-style exports)\n• Detects loops deterministically (identical repeat, ping-pong, retry-without-progress) via bundled Node scripts\n• Classifies failures into five buckets: tool selection, arguments, execution, orchestration, recovery policy\n• Writes a markdown triage report with targeted fix recommendations\n• Scaffolds regression tests (pytest / vitest templates)\n\nNo MCP servers, no external API calls, no npm dependencies. \n\nRequires Node.js 18+ for local scripts only. MIT licensed.","developer":null,"category":null,"reported_installs":null,"active":true,"first_seen_at":"2026-10-03T22:13:33.942Z","last_seen_at":"2026-10-04T12:03:33.240Z","last_changed_at":"2026-10-03T22:13:33.942Z","platform":"claude","sources":["Community"],"works_with":["Claude Code"],"marketplace_url":null,"repository_url":"https://github.com/WasayAbid/agent-trace-triage.git","web_data":{},"github_data":{},"catalog_sources":{"community":[{"name":"agent-trace-triage","source":{"sha":"03285a0b6fa6a9338782fe631e1966d68f9a2aba","url":"https://github.com/WasayAbid/agent-trace-triage.git","source":"url"},"homepage":"https://github.com/WasayAbid/agent-trace-triage","description":"Agent Trace Triage is a Claude Code plugin for developers building agentic AI systems (LangGraph, CrewAI, custom tool loops, MCP agents).\n\nWhen an agent run fails, stuck in a loop, wrong tool, or burning tokens. the user runs /agent-trace-triage with a JSON/JSONL trace file. The plugin:\n\n• Parses traces (JSONL, OpenAI tool_calls, partial LangSmith/OTEL-style exports)\n• Detects loops deterministically (identical repeat, ping-pong, retry-without-progress) via bundled Node scripts\n• Classifies failures into five buckets: tool selection, arguments, execution, orchestration, recovery policy\n• Writes a markdown triage report with targeted fix recommendations\n• Scaffolds regression tests (pytest / vitest templates)\n\nNo MCP servers, no external API calls, no npm dependencies. \n\nRequires Node.js 18+ for local scripts only. MIT licensed."}]}}