trace

Display chronological event timelines and aggregate summaries from Claude Code sessions.

Updated Feb 15, 2026
One-click install
npx skills add https://github.com/leejaedus/pepcode --skill trace-leejaedus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trace
Source: https://github.com/leejaedus/pepcode/tree/main/skills/trace
Command: npx skills add https://github.com/leejaedus/pepcode --skill trace-leejaedus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Skill provides real-time visibility into how hooks, keywords, skills, agents, and tools interact during a Claude Code session, enabling rapid debugging and optimization of multi-agent orchestration.

Core Features & Use Cases

  • Timeline: Show the chronological event sequence for the current or latest session.
  • Summary: Provide aggregate statistics such as hook fire counts, keywords detected, activated skills, mode transitions, and tool latency.
  • Use Case: Debug sessions where agents misbehave or performance bottlenecks occur, and optimize the orchestration flow.

Quick Start

Run trace_timeline with no arguments to display the latest session timeline and then trace_summary to view aggregate statistics.

Frequently Asked Questions about trace

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

FAQPage Schema
How do I debug a multi-agent orchestration flow in Claude Code?

You can debug multi-agent orchestration by viewing a chronological timeline and aggregate summary that audits interactions between hooks, keywords, skills, agents, and tools during a Claude Code session.

What is an agent flow trace timeline used for?

An agent flow trace timeline shows the chronological event sequence of a session, enabling rapid debugging and optimization of performance bottlenecks when agents misbehave.

How do I view aggregate statistics like hook fire counts and tool latency for my agents?

You view aggregate statistics like hook fire counts and tool latency by generating a trace summary, which provides aggregate statistics for activated skills, mode transitions, and detected keywords.

Can I filter specific event types when debugging sessions?

Yes, you can filter specific event types when debugging sessions; the visualization supports filtering of event types and a last parameter to narrow down the displayed chronological timeline.

Do I need specific MCP tools to visualize agent flow?

Yes, visualizing agent flow requires the trace_timeline and trace_summary MCP tools to output the chronological timeline and aggregate summary.

What is the best way to audit interactions between hooks, keywords, and skills?

The best way to audit interactions between hooks, keywords, and skills is to display the latest session timeline followed by an aggregate summary, revealing how these components interact chronologically.