Agent Harness: A Local-First, Lightweight Harness for AI Agent Evaluations
Abstract
The agent evaluation harness for the common man — simple, transparent, and accessible. A local-first, language-agnostic framework for running and grading AI agent benchmarks including GAIA, Terminal-Bench, ARC-AGI, and more.
Fields of Study & Subjects
Artificial Intelligence Applications
Scientific Computing and Data Management
AI-based Problem Solving and Planning
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 9 February 2026
- Language
- EN