Better Harness
A coding agent workflow analyzer that turns project and session evidence into prioritized findings, with support for Claude Code, Codex and Qoder.
Better Harness helps developers and teams assess how coding agents carry out work. It examines the workflow around a code change to find unclear goals, missing validation and delivery checks that a diff review can miss.
- Workflow assessment compares project instructions, specifications and reusable skills with evidence from tests, lint checks and supported agent sessions.
- Prioritized repairs ties each finding to supporting evidence, an expected outcome, a bounded repair and acceptance checks.
- Delivery tracing provides a read-only Harness Inspector that connects product intent with agent activity, sessions, files and commits.
- History comparisons shows changes across recorded workflow assessments while distinguishing trends from proof that a change improved results.
It runs through an existing coding agent, with support for Claude Code, Codex Desktop and CLI, Qoder Desktop and CLI, and GitHub Copilot CLI. Qoder Desktop includes it. Session evidence coverage depends on the host; Cursor installation is unavailable.
Better Harness is open source under the MIT License. Its reports distinguish evidence that a safeguard exists from evidence that an agent used it, and keep missing evidence visible.
