agentleFS
ccoalmccoalm/ccl-skills/skills/grill-me/SKILL.md

grill-me

Skill6 starsChanged 3 months agoApache-2.0View on GitHub

grill-me / 访谈 / 一问一答拷问(一次问一个)/ 压力测试方案 / stress-test a plan — lightweight one-question-at-a-time interview to challenge a plan, design, API shape, data model, or feature direction before implementation. Skip 拷问用的问题池·推荐默认值等材料,以及拷问后的结论整理(不是逐问过程)→ requirement-intent;code-level YAGNI/delete/adversarial review of written code → `product-rd-workflow`'s independent-review gate; full delivery/spec/plan authoring → product-rd-workflow; tests → testing-strategy; implementation → stack/dev skill; process lesson extraction → skill-extraction-workflow.

193 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use brief-to-tasks, writing-beats and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • julianoczkowskiSkill

    brief-to-tasks

    Break a design brief into an ordered checklist of independently buildable tasks using vertical slices. Saves as a markdown checklist. Use when user wants to break down work, create tasks from a brief, plan implementation order, or mentions "tasks" or "breakdown".

    Also in 7 projects that use this

  • LesaniSkill

    writing-beats

    Shape an article as a journey of beats, choose-your-own-adventure style. The user picks a starting beat from the raw material, you write only that beat, then offer options for where to pivot next, beat by beat, until the article reaches a natural end. Use when the user has raw material and wants to assemble it as a narrative rather than an argument.

    Also in 66 projects that use this

  • simo97Skill

    request-refactor-plan

    Use this skill when user wants to create a refactor plan.

    Also in 43 projects that use this

  • simo97Skill

    migrate-to-shoehorn

    Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.

    Also in 71 projects that use this

  • LesaniSkill

    writing-shape

    Take a markdown file of raw material and shape it into an article through a conversational session — drafting candidate openings, growing the piece paragraph by paragraph, arguing about format (lists, tables, callouts, quotes) at each step. Use when the user has a pile of notes, fragments, or a rough draft and wants help turning it into something publishable.

    Also in 65 projects that use this

  • liufeng1976Skill

    setup-ts-deep-modules

    Wire dependency-cruiser into a TypeScript repo so each package is a deep module — implementation hidden in subfolders, reachable only through its entry-point files. User-invoked.

    Also in 42 projects that use this

  • +2 more from simo97

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 177 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this when the user explicitly asks to grill, interview, challenge, or stress-test a plan before implementation.

    line 8
  2. Distinctive:

    Use for explicit requests such as:

    line 14
  3. Distinctive:

    before implementation, when several product or technical decisions depend on each other.

    line 19
  4. Very specific:

    Do not auto-trigger this from generic “帮我规划 / scope 一下 / 写个方案”.

    line 21
  5. Very specific:

    Test layers, fixtures, coverage, regression, CI gate design → testing-strategy.

    line 28
  6. Distinctive:

    Ask one blocking question at a time.

    line 37
  7. Distinctive:

    If the answer depends on existing repo facts, inspect the relevant files first.

    line 49
  8. Distinctive:

    Ask exactly one question with the recommended answer and why.

    line 55
  9. Distinctive:

    After the answer, update the assumption and choose the next dependent question.

    line 56
  10. Distinctive:

    Stop when the direction is clear enough to route to the next owner or when the user rejects the premise.

    line 57
  11. Distinctive:

    Avoid bundled questions. Avoid long lectures.

    line 59
  12. Distinctive:

    When the interview reaches a stable stop, return:

    line 63

“Common” is measured: kept by at least 6 in 10 of the 177 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

grill-me — 轻量方案拷问

Use this when the user explicitly asks to grill, interview, challenge, or stress-test a plan before implementation.

This is a narrow pre-implementation interview. It clarifies the next blocking decision; it does not write the plan, own delivery gates, or replace the relevant owner workflow.

When to use

Use for explicit requests such as:

  • grill-me / grill me;
  • 访谈、拷问、追问一个方案;
  • 压力测试方案、设计、API shape、数据模型、权限模型、工作流;
  • before implementation, when several product or technical decisions depend on each other.

Do not auto-trigger this from generic “帮我规划 / scope 一下 / 写个方案”. Those route to the owning workflow first, usually product-rd-workflow.

Owner boundaries

  • Full requirement shaping, spec, plan, lifecycle gates, or multi-step delivery → product-rd-workflow.
  • Risk tags, gate selection, rollout or review requirements → feature-risk-router.
  • Code-level YAGNI, delete-code, overengineering, or adversarial review of already-written code → product-rd-workflow's independent-review gate (no dedicated CCL delete-code/YAGNI skill; the review gate owns it).
  • Test layers, fixtures, coverage, regression, CI gate design → testing-strategy.
  • Bug/root-cause diagnosis → defect-diagnosis.
  • Implementation → the relevant stack/dev skill or bounded executor.
  • Reusable process/skill lesson extraction → skill-extraction-workflow.

grill-me may feed those owners by resolving one unclear decision first. It must hand back once the user asks for a deliverable artifact, implementation, formal review, or test plan.

Core contract

Ask one blocking question at a time.

Each turn must include:

  1. Target: the plan/design being challenged.
  2. Current assumption: the most important assumption currently visible.
  3. Question N: exactly one unresolved decision.
  4. Recommended answer: a concrete default.
  5. Why: short rationale and the risk if the answer is wrong.

Then wait for the user's answer before asking the next dependent question.

If the answer depends on existing repo facts, inspect the relevant files first. Do not ask the user to restate facts that are available in code, specs, runbooks, or local docs.

Workflow

  1. Frame the target. If the target is ambiguous, ask one clarifying question only.
  2. Identify the next decision that blocks meaningful progress.
  3. Ask exactly one question with the recommended answer and why.
  4. After the answer, update the assumption and choose the next dependent question.
  5. Stop when the direction is clear enough to route to the next owner or when the user rejects the premise.

Avoid bundled questions. Avoid long lectures. Do not produce a full PRD, architecture doc, implementation plan, or code unless the user explicitly switches workflows.

Closeout

When the interview reaches a stable stop, return:

Resolved direction
- <decision>: <answer>

Remaining risks
- <risk or unknown>

Recommended next step
- <owner workflow / implementation / no-op>

Tone

  • Direct, skeptical, and practical.
  • Push back on weak assumptions with evidence.
  • No flattery.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.