trace

Visualize agent flow traces and summaries to diagnose execution paths.

38.5k|3.5k|Updated Jan 9, 2026
One-click install
npx skills add https://github.com/Yeachan-Heo/oh-my-claudecode --skill trace-yeachan-heo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace
Source: https://github.com/Yeachan-Heo/oh-my-claudecode/tree/main/skills/trace
Command: npx skills add https://github.com/Yeachan-Heo/oh-my-claudecode --skill trace-yeachan-heo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Visualize agent flow traces and their summary to diagnose execution paths and improve debugging workflows.

Core Features & Use Cases

  • Timeline view of interactions between hooks, keywords, skills, agents, and tools in a session
  • Aggregate statistics such as hook fire counts, keywords detected, skills activated, and mode transitions
  • Use Case: When debugging an orchestration, generate a timeline for the latest session and filter by event type

Quick Start

Use the trace skill to display the latest session trace with the timeline and summary. Use the trace_timeline MCP tool to show the chronological event timeline, and the trace_summary MCP tool to show aggregate statistics.

Frequently Asked Questions about trace

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I visualize an agent flow trace to debug execution paths?

Multi-agent session traces visualize interactions between hooks, keywords, skills, agents, and tools in a timeline view to diagnose execution paths. Use the trace_timeline tool to display the chronological event timeline and trace_summary for aggregate statistics.

What is the best way to get aggregate statistics for a debugging session?

The best way to get aggregate statistics for a debugging session is using the trace_summary tool, which provides aggregate statistics such as hook fire counts, keywords detected, skills activated, and mode transitions to monitor multi-agent interactions.

Do I need external libraries to trace agent execution workflows?

No, you do not need external libraries to trace agent execution workflows. The trace functionality requires no external libraries beyond built-in tracing utilities to generate timelines and summaries for multi-agent sessions.

Can I filter the event timeline by specific interaction types during orchestration debugging?

Yes, you can filter the event timeline by event type during orchestration debugging. When debugging an orchestration, generate a timeline for the latest session and filter by event type to inspect specific hook, keyword, skill, agent, or tool interactions.

How do I retrieve the trace timeline for the last session only?

To retrieve the trace timeline for the last session only, use the trace_timeline MCP tool with last-session retrieval support. This displays the chronological event timeline specifically for the most recent multi-agent execution session.

Why does my multi-agent session trace show unexpected mode transitions?

Unexpected mode transitions in a multi-agent session trace can be diagnosed using the trace_summary tool, which provides aggregate statistics on mode transitions and insights into how hooks, keywords, skills, agents, and tools interacted during the session.