Skip to content
SkillCheatSheet Sign in

harness-eval

An AI agent skill from HKUDS/OpenHarness, scored 89 out of 100 by code and Verified.

89/ 100AVerified

Code score by group

  • Spec20/20
  • Description17/20
  • Procedure15/15
  • Done criteria13/15
  • Resources10/10
  • Scripts2/8
  • Safety7/7
  • Upkeep5/5

It loses the most points in Scripts (6), the group that checks whether scripts are called from the steps and fail with a clear error.

What the author says it does

This skill should be used when the user asks to "test the harness", "run integration tests", "validate features with real API", "test with real model calls", "run agent loop tests", "verify end-to-end", or needs to verify OpenHarness features on a real codebase with actual LLM calls.

This is the description in the skill's own SKILL.md, written by its author, not by SkillCheatSheet.

Repository
HKUDS/OpenHarness
Folder
.claude/skills/harness-eval
License
MIT
GitHub stars
16,200
SKILL.md lines
194
Other files
2

Add harness-eval to your agent

npx skills add HKUDS/OpenHarness --skill harness-eval

The command uses the open skills CLI. A skill is instructions and scripts your agent will follow, so read it first: open the folder on GitHub at the commit that was scored.

Labelled for

More skills from HKUDS/OpenHarness

  1. pr-merge

    This skill should be used when the user asks to "merge a PR", "review and merge pull requests", "integrate external contributions", "handle PR conflicts", "cherry-pick from a PR",…

    78B

All skills in HKUDS/OpenHarness

Scores come from a snapshot of 300 public GitHub repositories taken on 12 November 2026. Every score is the result of fixed checks run by code; no AI is involved. How scoring works.