eval-harness

80 / 100 · Low risk

Fine to try. Some of what it can do is expected: a coding skill usually runs build and test commands.

By affaan-m/ECC (GitHub) · coding · 300,508 stars · scanned at commit d2ef268

Eval-driven development (EDD) framework for AI coding sessions — define capability and regression evals before coding, grade with code-based, model-based, rule, or human graders, and track pass@k and pass^k reliability. Use when defining pass/fail criteria for agent tasks, measuring agent reliability, building regression suites for prompt or agent changes, or benchmarking across model versions.

Works with any agent that supports SKILL.md skills.

What this skill can do

Runs commands on your computer

The author does not mention this. It runs automatically when the plugin is installed. Expected for a coding skill.

Found in 47 places.

skills/eval-harness/SKILL.md, line 61grep -q "export function handleAuth" src/auth.ts && echo "PASS" || echo "FAIL"
skills/eval-harness/SKILL.md, line 134[Run each capability eval, record PASS/FAIL]
skills/eval-harness/SKILL.md, line 244node scripts/eval-harness.js example

Show 17 more

scripts/hooks/observe-runner.js, line 6const { spawnSync } = require('child_process');
scripts/hooks/observe-runner.js, line 63const probe = spawnSync(candidate, ['-c', ':'], {
scripts/hooks/observe-runner.js, line 131const result = spawnSync(shell, [toShellPath(observePath), phase], {
scripts/hooks/session-activity-tracker.js, line 13const { spawnSync } = require('child_process');
scripts/hooks/session-activity-tracker.js, line 346const result = spawnSync('git', args, {
scripts/lib/llm-summary.js, line 15const { spawnSync } = require('child_process');
scripts/lib/llm-summary.js, line 180const result = spawnSync('claude', ['--model', getLLMModel(), '-p'], {
scripts/lib/observer-sessions.js, line 5const { spawnSync } = require('child_process');
scripts/lib/observer-sessions.js, line 45const result = spawnSync('git', args, {
scripts/lib/utils.js, line 10const { execSync, spawnSync } = require('child_process');
scripts/lib/utils.js, line 491const result = spawnSync('where', [cmd], { stdio: 'pipe' });
scripts/lib/utils.js, line 494const result = spawnSync('which', [cmd], { stdio: 'pipe' });
scripts/lib/utils.js, line 528const result = execSync(cmd, {
.mcp.json, line 4"command": "npx",
hooks/README.md, line 295"command": "node -e \"let d='';process.stdin.on('data',c=>d+=c);process.stdin.on('end',()=>{const i=JSON.parse(d);const p=i.tool_input?.file_path||'';if(/\\.py$
hooks/codex-hooks.json, line 10"command": "node -e \"if(!process.env.PLUGIN_ROOT)throw new Error('Missing Codex PLUGIN_ROOT');process.env.CLAUDE_PLUGIN_ROOT=process.env.PLUGIN_ROOT;const p=re
hooks/hooks.json, line 9"command": "node -e \"const p=require('path');const r=(function(){var p=require('path'),f=require('fs'),o=require('os');var e=process.env.CLAUDE_PLUGIN_ROOT;if(

and 27 more places

Downloads and installs other software

The author does not mention this. It runs automatically when the plugin is installed.

Found in 2 places.

scripts/lib/package-manager.js, line 29execCmd: 'pnpm dlx',
.mcp.json, line 4"command": "npx",

Other copies in this repository

affaan-m/ECC has 7 more folders with a skill named eval-harness. This page shows the lowest-scoring copy.

.agents/skills/eval-harness 80/100 · Low risk
docs/es/skills/eval-harness 80/100 · Low risk
docs/tr/skills/eval-harness 80/100 · Low risk
docs/ja-JP/skills/eval-harness 80/100 · Low risk
docs/ko-KR/skills/eval-harness 80/100 · Low risk
docs/zh-CN/skills/eval-harness 80/100 · Low risk
docs/zh-TW/skills/eval-harness 80/100 · Low risk

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'eval-harness' ] || [ -L ~/.agents/skills/'eval-harness' ]; then echo 'A skill folder named eval-harness already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/affaan-m/ECC' "$d" && git -C "$d" checkout --quiet 'd2ef2680594f9a17526a6e1a2e18b3d242c5d17b' -- 'skills/eval-harness' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/eval-harness' ~/.agents/skills/'eval-harness' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'eval-harness' ~/.claude/skills/'eval-harness' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for affaan-m/ECC shows the score of its lowest-scoring skill.

SkillTrust badge for affaan-m/ECC

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/affaan-m/ECC.svg)](https://zalavijaysinh.github.io/skilltrust/repos/affaan-m/ECC/)

Score history

Date	Version	Score
2026-11-08	`d2ef268`	80
2026-11-06	`b38be6a`	80

Also found in

Identical copies of this skill exist at:

affaan-m/everything-claude-code/skills/eval-harness