session-report

Generate explorable HTML reports of Claude Code session usage from transcripts.

Updated May 11, 2026
One-click install
npx skills add https://github.com/AvaTar-ArTs/my-supremepowers --skill session-report-avatar-arts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: session-report
Source: https://github.com/AvaTar-ArTs/my-supremepowers/tree/main/qwen_skills/session-report
Command: npx skills add https://github.com/AvaTar-ArTs/my-supremepowers --skill session-report-avatar-arts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill generates an explorable HTML report of Claude Code session usage from transcripts, enabling quick insights into token consumption, cache behavior, subagents, and prompts to support optimization and auditing.

Core Features & Use Cases

  • Holistic reporting: Consolidates overall token usage, per-project trends, per-subagent-type breakdowns, per-skill metrics, and top prompts into a single HTML report.
  • Drill-down analysis: Allows exploration by project, subagent, and prompt to identify hotspots, inefficiencies, and opportunities for optimization.
  • Template-driven rendering: Uses a bundled HTML template to ensure consistent, shareable reports for teams and stakeholders.

Quick Start

Run the bundled analyze-sessions.mjs to generate the JSON report and render it with template.html to produce the final HTML output.

Frequently Asked Questions about session-report

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

FAQPage Schema
How do I generate an HTML report from Claude Code session transcripts?

To generate an HTML report from Claude Code session transcripts, run the bundled analyze-sessions.mjs script to aggregate token counts and cache events, then render the output with template.html to produce a self-contained HTML file.

Can I analyze token usage and cache events across multiple Claude Code projects?

Yes, you can analyze token usage and cache events across multiple projects by scanning transcripts stored in ~/.claude/projects, aggregating per-project trends, per-subagent-type breakdowns, and per-skill metrics into a single report.

What is the best way to audit Claude Code subagent and prompt metrics?

The best way to audit Claude Code subagent and prompt metrics is to consolidate session transcripts into an explorable HTML report, allowing drill-down analysis by project, subagent, and top prompts to identify inefficiencies.

Does session reporting require any specific environment setup or dependencies?

Session reporting requires no external dependencies, but it does need Claude Code transcripts located in ~/.claude/projects and the bundled analyze-sessions.mjs script alongside template.html to render the final HTML output.

Why are my Claude Code session reports not showing per-skill metrics?

Per-skill metrics require session transcripts stored under ~/.claude/projects that contain skill invocation data; the analyze-sessions.mjs script aggregates this data, so missing metrics indicate empty transcripts or incomplete session logs.

Can I share the generated Claude Code session report with my team?

You can share the generated report because the template-driven rendering produces a self-contained, ready-to-render HTML file that consolidates token consumption, cache behavior, and subagent breakdowns for stakeholders.