session-analyzer

Parse JSONL session files to compute usage metrics and identify cost drivers.

Updated Jan 24, 2026
One-click install
npx skills add https://github.com/reubenjohn/reeve-desk --skill session-analyzer-reubenjohn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-analyzer
Source: https://github.com/reubenjohn/reeve-desk/tree/main/.claude/skills/session-analyzer
Command: npx skills add https://github.com/reubenjohn/reeve-desk --skill session-analyzer-reubenjohn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyze Claude Code sessions to extract metrics for retrospection and budget tracking, enabling visibility into session costs, usage patterns, and feedback signals.

Core Features & Use Cases

  • Metrics extraction: parse JSONL session files to compute message counts, tool usage, input/output tokens, and durations.
  • Retrospection support: identify cost drivers and usage patterns for better budgeting and process improvement.
  • Session debugging: analyze specific sessions to review performance and user feedback.

Quick Start

Run the session-analyzer on today's Claude Code sessions to generate a metrics report.

Frequently Asked Questions about session-analyzer

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

FAQPage Schema
How do I analyze Claude Code session metrics for retrospection and budgeting?

To analyze Claude Code session metrics for retrospection, you can parse JSONL session files and the sessions index to compute message counts, tool usage, input/output tokens, and durations into a summarized metrics report.

What session usage patterns and cost drivers can I extract from JSONL files?

From JSONL files, you can extract usage patterns and cost drivers by parsing input/output tokens, tool usage frequencies, session durations, and user feedback signals to identify high-cost activities.

Can I debug a specific Claude Code session by reviewing its token usage and timing data?

Yes, you can debug a specific Claude Code session by analyzing its JSONL file to review performance metrics, token consumption, tool usage, and user feedback signals.

Does session-analyzer work without external dependencies for monitoring tool usage?

Yes, session-analyzer operates without external dependencies, directly reading JSONL session files and the sessions index to monitor tool usage, tokens, and feedback signals autonomously.

What is the best way to track daily Claude Code session costs and feedback signals?

The best way to track daily Claude Code session costs is to run retrospection analysis on JSONL session files, summarizing token counts, durations, and feedback signals into a metrics report.