genshijin-stats

Report real token usage and estimated savings from Claude Code session logs.

304|10|Updated Apr 6, 2026

Search Tags:#Claude Code #compression #cost optimization #token usage #session analytics #prompt tuning #lifecycle reporting

Ready to run

npx skills add https://github.com/InterfaceX-co-jp/genshijin --skill genshijin-stats

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: genshijin-stats
Source: https://github.com/InterfaceX-co-jp/genshijin/tree/main/skills/genshijin-stats
Command: npx skills add https://github.com/InterfaceX-co-jp/genshijin --skill genshijin-stats

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes hooks (resource) components.

What problem does it solve?

Tokens are expensive, but it’s hard to know how much you actually saved in the current chat session.

Core Features & Use Cases

Real-time session token stats: Displays the current session’s real token usage and the estimated reduction.
Claude Code log-based, no model guesswork: Reads directly from the session log rather than relying on AI estimation.
Shareable summaries and lifetime views: Supports one-line sharing and lifetime aggregation with optional time windows.
Use case: Before/after comparing different compression modes while you iterate on prompt style and verify cost impact.

Quick Start

In your Claude Code chat, run /genshijin-stats to show the current session token usage and estimated savings.

Frequently Asked Questions about genshijin-stats

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

FAQPage Schema

How do I check real token usage in a Claude Code session?▼

You can check real token usage in a Claude Code session by triggering the /genshijin-stats command, which reads directly from session logs to display actual token consumption and estimated savings without relying on AI estimation.

Can I compare token costs before and after prompt tuning?▼

Yes, you can compare token costs before and after prompt tuning by running the stats command across iterations, which evaluates compression-mode comparisons and provides cost feedback across your chat time window.

Does the token savings report rely on model guesswork?▼

No, the token savings report does not rely on model guesswork because it reads actual token usage data directly from the Claude Code session logs, ensuring accurate cost feedback while the model performs no numeric computation.

What is the best way to view lifetime token savings for Claude Code?▼

The best way to view lifetime token savings for Claude Code is by using the session analytics command, which supports lifetime aggregation with optional time windows to track cost optimization across multiple sessions.

How do I share my Claude Code session analytics summary?▼

You can share your Claude Code session analytics summary by triggering the reporting hook, which generates preformatted stats with a one-line sharing format that highlights real token usage and estimated reductions.