agentleFS
Sign inStart free
BSSE23047BSSE23047/nauraiz-dev/skills/grill/SKILL.md

grill

Skill2 starsChanged 14 days agoMITView on GitHub

Ask the owner the few questions that change what gets built, before the plan: open decisions, shared paths, required approvals. Use from start-task or when asked to clarify a task. Not for small fixes or what the repo already answers.

87 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use write-skill, root-cause 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.
  • NoaiRoxSkill

    write-skill

    Use when creating a new skill, editing an existing skill, or verifying a skill works before deployment.

    Also in 5 projects that use this

  • Ingramradical235Skill

    root-cause

    Structured root cause analysis using WHERE (MECE decomposition) then WHY (causal structure diagrams) pipeline. Includes coin-flip detection, logical-leap detection, and vicious cycle intervention. Use when buffer enters YELLOW/RED, when metrics stall, or when Actions execute but don't produce results.

    Also in 4 projects that use this

  • Neko-Catpital-LabsSkill

    prove-it

    Shared evidence rule for investigating, diagnosing, debugging, proving, or explaining what happened or why for any problem or situation: do not assert something is fixed, working, passing, merged, or running unless you personally observed it this turn and can show what you saw; do not assert it was caused by X until a repro and controlled isolation prove that cause.

    Also in 4 projects that use this

  • sahinSkill

    execute

    Also in 5 projects that use this

  • assisjpSkill

    file-pr

    Prepare and open or update an authorized pull request with a concise problem-first description and verifiable evidence. Use when the user asks to create, file, open, or update a PR. Do not use merely to review a diff or when publication is not authorized.

    Also in 3 projects that use this

  • Arman-KudaibergenovSkill

    write-plan

    Этот скилл MUST быть вызван когда есть design.md и нужно разбить доработку на атомарные задачи (2-5 мин) с точными путями файлов и критериями проверки в tasks.md. SHOULD также вызывать как промежуточный шаг между brainstorm и subagent-dev. Do NOT использовать для обсуждения идеи — используй brainstorm; для выполнения задач — используй subagent-dev.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 69 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask only questions the repo and the issue cannot answer and where a wrong guess changes what gets built.

    line 9
  2. Very specific:

    Ask with AskUserQuestion: one call, multiple choice, the recommended option first and labelled "(Recommended)".

    line 32

“Common” is measured: kept by at least 6 in 10 of the 69 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

Ask only questions the repo and the issue cannot answer and where a wrong guess changes what gets built. Search first: the issue and its comments, the code (Grep, Glob), the project's .claude/ and its plan template (plan.template in .claude/nauraiz-dev.json, with .claude/nauraiz-dev.local.json over it). Never ask what you could have looked up.

Must ask, before the plan, when any of these holds

  • The issue has a "Decisions needed" or "Open decision" section, or contains the words "decide", "confirm" or "TBD".
  • The work touches a path that matches a sharedPaths pattern (gitignore syntax) in the config, or the plan template's kind field offers a value containing shared (such as shared-seam) and that is the kind this work is.
  • The plan template has any *_approval field: the owner must give that approval.

Your own judgement may add a question; it never removes a must-ask. If nothing holds and nothing material is open, say "No questions: <one-line reason>" and return.

How to ask

  • At most 3 questions per round, normally one round. Keep the ones with the highest cost if guessed wrong.
  • Each question gives the options, your recommendation, and the cost if the recommendation is wrong (what would have to be redone).
  • Ask with AskUserQuestion: one call, multiple choice, the recommended option first and labelled "(Recommended)". If it is unavailable, errors, or returns no answer, follow the headless rule below.
  • An approval question asks for the approval itself. Record it only on a yes, with the date and the owner's words; otherwise it stays pending owner confirmation.
  • The owner's answers are decisions: hold them until step 8 writes the plan (start-task), which puts them under ## Decisions with the date and the owner's words. "Decisions to confirm" is only for choices the owner did not state.

Headless: no human to answer

You are headless when the request says the run is unattended, headless or non-interactive (for example claude -p), or AskUserQuestion is not available, errors or returns no answer. Then do not ask and do not wait: write each question with its recommended answer under "Decisions to confirm" in the plan (in the reply if there is no plan yet), and proceed on the recommendation. A recommendation is never recorded as the owner's decision or approval.

After the round

  • Play back in two to four lines what will be built, with the answers.
  • Hand back to start-task: it runs the next phase it routed (the Design phase for large tasks, the plan for normal ones). Called on its own, stop after the play-back.

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.

Check your setup, free