菜单
Run Mini SWE-bench style harness regression tasks and A/B comparisons to measure harness improvement objectively.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Build a lightweight skill index and load full skill instructions only on demand to reduce startup context pressure.
Select and run a multi-agent workflow pattern for work that exceeds one agent's reliable scope. Use for parallel research, independent reviews, cross-domain changes, or high-risk implementation planning.
Select and run a multi-agent workflow pattern for work that exceeds one agent's reliable scope. Use for parallel research, independent reviews, cross-domain changes, or high-risk implementation planning.
Use composable middleware behaviors for provider/tool execution: retry, caching, timeout, telemetry, and budget guards.
Use this skill whenever the agent makes a mistake, the user observes an avoidable failure, a pattern recurs, or someone says "the agent keeps doing X". Files an "Engineer the Harness" entry — Mitchell Hashimoto's discipline: every failure becomes a permanent prevention mechanism. Always invoke this instead of just fixing the immediate symptom.
Use composable middleware behaviors for provider/tool execution: retry, caching, timeout, telemetry, and budget guards.
| name | benchmark-suite |
| description | Run Mini SWE-bench style harness regression tasks and A/B comparisons to measure harness improvement objectively. |
| allowed-tools | Read, Bash(node .harness/scripts/bench-runner.mjs:*), Bash(node .harness/scripts/bench-compare.mjs:*) |
| suggested-turns | 6 |
Use this when evaluating whether a harness change improved or regressed behavior.
node .harness/scripts/bench-runner.mjs --variant=current
node .harness/scripts/bench-runner.mjs --variant=candidate
node .harness/scripts/bench-compare.mjs
### Benchmark Suite
### Tasks: <n>
### Pass rate: <percent>
### Avg score: <score>
### A/B delta: <delta or n/a>