See it in action
Ask /better-harness to review the current task and its surrounding project Harness. The report keeps missing evidence explicit and turns supported gaps into prioritized findings with an impact, expected output, scoped repair, and acceptance checks.
Open the complete self-contained English HTML report
The animation replays historical Harness reports. It shows recorded trends, not causal proof of improvement.
How Better Harness works
Better Harness combines feedforward guides (AGENTS.md, specs, Skills, acceptance criteria) with feedback sensors (linters, tests, Hooks, review agents), and evaluates five parts of delivery—the Agent Work Loop:
Task Understanding
Does the agent know the goal and what “done” means?
Controlled Execution
Is the work on supported, repeatable paths?
Change Validation
Is there evidence the change actually works?
Reliable Delivery
Does AI speed bypass quality checks or acceptance?
Learning Capture
Does the next task benefit from this one?
Three evidence domains stay independent until unified analysis; every result retains a visible evidence source, owner, and validation route.
Quick start
Pick your coding agent—you can be looking at your first report in minutes. Once installed, ask for the host's durable report:
/better-harness review this project's AI coding workflow and generate a report