analyzing-cc-sessions

Parse JSONL transcripts to extract metrics and diagnose subagent behavior.

6|1|Updated Dec 8, 2025

Search Tags:#debugging #metrics #subagents #analyze #jsonl #session-logs

Ready to run

npx skills add https://github.com/vladolaru/claude-code-plugins --skill analyzing-cc-sessions

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: analyzing-cc-sessions
Source: https://github.com/vladolaru/claude-code-plugins/tree/main/plugins/pirategoat-tools/skills/analyzing-cc-sessions
Command: npx skills add https://github.com/vladolaru/claude-code-plugins --skill analyzing-cc-sessions

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Claude Code session data is noisy and scattered across main JSONL transcripts and per-subagent logs. This Skill provides a structured approach to parse, correlate, and extract actionable metrics from raw CC session data to diagnose performance, behavior, and tool-use patterns.

Core Features & Use Cases

Parse and correlate main sessions with subagent dispatch files to reveal end-to-end workflows.
Extract metrics such as tool usage, durations, and findings to support reliability and efficiency analyses.
Debug subagent dispatches and perform cross-session comparisons to identify bottlenecks and improvement opportunities.

Quick Start

Run the session analyzer on a target project to generate metrics for a chosen agent type.

Frequently Asked Questions about analyzing-cc-sessions

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

FAQPage Schema

How do I extract metrics from Claude Code JSONL session logs?▼

To extract metrics from Claude Code JSONL session logs, run analysis scripts that parse raw transcripts and subagent files to surface tool usage, durations, and actionable findings for your projects.

What is the best way to debug subagent dispatch behavior in Claude Code?▼

Debugging subagent dispatch behavior involves mapping main JSONL sessions to per-subagent log files to diagnose tool-use patterns, identify bottlenecks, and reveal end-to-end workflows across project directories.

Can I compare Claude Code session metrics across multiple project directories?▼

Yes, you can perform cross-session comparisons across multiple project directories by applying robust JSONL parsing to map main sessions to subagents, generating metrics to identify bottlenecks and improvement opportunities.

Do I need specific dependencies to analyze raw Claude Code session transcripts?▼

No specific dependencies are required to analyze raw Claude Code session transcripts; the Skill provides standalone analysis scripts to parse JSONL logs and subagent files directly within your environment.

Why are my Claude Code subagent logs scattered and hard to correlate with main sessions?▼

Subagent logs are scattered across separate per-subagent files, but this Skill provides a structured approach to parse and correlate main JSONL transcripts with dispatch files to reveal complete end-to-end workflows.