benchmark-audit

Official

Audit benchmark quality and validity.

Authorsourcegraph
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill ensures the quality and reliability of benchmark suites by auditing them against the Agent Benchmark Criteria (ABC) framework, identifying potential issues before they impact evaluations.

Core Features & Use Cases

  • Comprehensive Auditing: Evaluates Task Validity, Outcome Validity, and Reporting dimensions of benchmark tasks.
  • Detailed Reporting: Provides letter grades (A-F) and specific issue identification for each dimension.
  • Use Case: Before launching a new set of benchmarks for AI code generation, use this Skill to perform a pre-flight check, ensuring all tasks are well-defined, verifiable, and reportable, catching critical flaws early.

Quick Start

Use the benchmark-audit skill to audit all benchmark suites and display the results in a table format.

Dependency Matrix

Required Modules

None required

Components

scriptsreferences

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: benchmark-audit
Download link: https://github.com/sourcegraph/CodeScaleBench/archive/main.zip#benchmark-audit

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.