Skills
.Work. You
Rest

cross-harness

Community

Benchmark LongMemEval across Codex, Gemini, and OpenCode harnesses.

Software Engineering#benchmarking#QA#Codex#Gemini#OpenCode#LongMemEval
Authortmuskal
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill automates the benchmarking of LongMemEval across different harnesses (Codex, Gemini, OpenCode) and integrates their results into a single scorecard.

Core Features & Use Cases

  • Cross-Harness Benchmarking: Execute LongMemEval tasks on multiple harnesses simultaneously.
  • Result Aggregation: Combine and compare results from different harnesses in a unified format.
  • Use Case: If you want to compare the performance of your model across Codex, Gemini, and OpenCode, this Skill can help you automate the process and generate a comprehensive report.

Quick Start

Use the cross-harness skill to benchmark your model across different harnesses with the command: /cross-harness:run <action> --harness <harness> --run-id <run-id>.

Dependency Matrix

Required Modules

None required

Components

scripts

💻 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: cross-harness
Download link: https://github.com/tmuskal/arc-agi-benchmarker/archive/main.zip#cross-harness

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 620,000+ vetted skills library on demand.