skills.rest
guyman-tr
guyman-trCOMMUNITY
Explore all skills by this author →
GitHub

session-bench

Measure and grade Claude Code sessions by token usage and context growth.

Updated Aug 23, 2026
Search Tags:#benchmarking#Claude Code#code session#token cost#efficiency analysis
One-click install
npx skills add https://github.com/guyman-tr/Databricks_Knowledge --skill session-bench
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: session-bench
Source: https://github.com/guyman-tr/Databricks_Knowledge/tree/main/.claude/skills/session-bench
Command: npx skills add https://github.com/guyman-tr/Databricks_Knowledge --skill session-bench

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, and includes scripts (resource) components.

What problem does it solve?

This Skill allows users to measure and grade the cost and efficiency of Claude Code sessions, helping to identify improvements in context management and token usage.

Core Features & Use Cases

  • Session Benchmarking: Analyze the cost of a Claude Code session, including tokens used and context growth.
  • Comparison Against Baseline: Compare the performance of a session against a previously set baseline.
  • Efficiency Analysis: Evaluate the efficiency of a session and suggest improvements.

Quick Start

Measure the cost of your last session by running: python tools/session-bench/measure_session.py --save.

Frequently Asked Questions about session-bench

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

FAQPage Schema
How do I measure token cost and context growth in a Claude Code session?▼

Token cost benchmarking evaluates your Claude Code session transcripts to measure token usage and context growth, then grades efficiency by comparing current session performance against a previously saved baseline.

How do I benchmark my Claude Code session against a previous baseline?▼

You benchmark against a baseline by saving your initial session metrics with a flag, then running subsequent session analyses to compare current token usage and context growth against that saved performance baseline.

Do I need Python to analyze Claude Code session transcripts for token efficiency?▼

Yes, Python is required to execute the session analysis scripts that parse Claude Code transcripts, calculate token usage metrics, and generate efficiency grades against your established baseline.

What is the best way to analyze token usage efficiency in Claude Code?▼

The best way to analyze token usage efficiency is to benchmark Claude Code sessions by measuring token consumption and context growth, then comparing those metrics against a defined baseline to identify improvement areas.

Can I get suggestions for improving context management after benchmarking a code session?▼

Yes, after benchmarking your code session the tool provides efficiency analysis that evaluates token usage and context growth, offering feedback and suggestions for improving context management in future sessions.

Related Skills

Explore More Software Engineering Skills→
sejassejas
2

context-window-statusline

Generate and install a customizable Claude Code status line from JSON session data.

Community
Intermediate
ZESCODEZESCODE

ZES-context-manager

Monitor token consumption and suggest strategic compaction intervals for multi-agent sessions.

Community
Advanced
Manvendra08Manvendra08
1

token-efficiency

Reduce token consumption and context overflow in LLM interactions.

Community
Basic
clinecline
19

session-report

Analyze Claude Code JSONL transcripts to generate interactive HTML usage reports.

Official
Intermediate