菜单
Build and harden a skill with evals — interview to design its eval tasks, then run, measure, and iterate. Use when the user wants to create or improve a skill's eval, or run the create → test → improve loop for a skill.
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。
Measure a skill's reliability — run it k times for a pass@k score, design or interpret its eval, or compare it against the base agent. Use when the user wants to run, design, or interpret a skill's eval, or write an .eval.yaml spec.
Save a short note to a file. Use when asked to record, save, or jot down a note.
Send an operational alert. Use when asked to send an alert, page, or notification.
Fetch data with a strict retry budget. Use when asked to fetch data that may be unavailable.
Convert a title into a URL slug. Use when asked to slugify a title or make a URL slug.
Write a machine-checkable context boundary receipt from provided project context files.
基于 SOC 职业分类
| name | grill-skill |
| description | Build and harden a skill with evals — interview to design its eval tasks, then run, measure, and iterate. Use when the user wants to create or improve a skill's eval, or run the create → test → improve loop for a skill. |
| allowed-tools | Bash, Read, Write, Edit |
Interview the user to design a skill's eval, then loop run → measure → improve until it ships. Requires caliper (pipx install caliper-eval if missing). Commands, spec skeleton, and expect/assert guidance: REFERENCE.md.
/grill-skill [path] — optional path to a SKILL.md.
SKILL.md in the cwd; if found, confirm before proceeding, else ask where it is.Read the SKILL.md. Summarize what it does, when it triggers, and what a successful run looks like. Ask the user to confirm your reading. Wait for confirmation before continuing.
Look for *.eval.yaml beside the SKILL.md (try <dir-name>.eval.yaml first).
Interview one question at a time and wait for each answer. Never invent the user's answers or write the spec before interviewing.
Elicit three tasks, one question at a time:
Turn each answer into a task: a realistic prompt, an observable expect, and an assert when the outcome is checkable (see REFERENCE.md). Show the proposed YAML and confirm before writing.
Write the spec beside SKILL.md, named <dir-name>.eval.yaml, with skill.path: ./SKILL.md and claude-code as the default backend for both skill and judge unless the SKILL.md targets another.
Read the existing spec and report its tasks. Ask what behaviors are missing or under-tested before proposing or writing anything — even if the user only asked you to inspect it, report first, then ask. Sharpen each gap into a task, show it, and confirm before writing it in.
Validate the spec, then run at k=1 (commands in REFERENCE.md). Show the results. Fix any harness or config error (not a task failure) before asking the user what to do next.
Ask whether to iterate or finish.
SKILL.md, re-run at k=3 and show results. Loop back.--baseline run to prove the skill beats the raw agent, then remind the user to commit SKILL.md and the .eval.yaml together.