agentleFS
mantaray0mantaray0/crew/skills/roast-me/SKILL.md

roast-me

Skill1 starsChanged 5 months agoMITView on GitHub

Clarify and challenge a raw idea in rounds of questions, each with a recommended answer, until a Spec can be written. Use in /crew:brief, or when the user says "roast me" or asks to challenge an idea, plan or assumption.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. Pasquale-Favella/watchtower★ 1same name
    2. immoJUMP/immo-agent-skills★ 0same name
    3. vanities/skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Challenge, don't just collect: Every round, name the assumption the idea leans on hardest and test it: what breaks if it is false, what the …

    line 17

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Roast-Me

Turn a raw idea into a decided Spec (goal, requirements, acceptance criteria, out of scope) by walking its decision tree with the user and pushing on the weak spots.

Rounds

  1. Look before you ask. Anything the code, the configs, PROJECT.md or the web can answer, find out yourself; a wide search can run in an Explore sub-agent while you ask what doesn't hinge on it.
  2. Ask everything that is askable now. A question is askable once every decision it hangs on is settled: trunk before leaves. Put all askable questions into one round; a question whose options depend on an answer still open in this round waits for the next.
  3. Each question carries a recommendation with a one-line why, so the user can confirm it at a glance. A recommendation may be "drop this".
    • Closed choices with 2–4 options: AskUserQuestion, up to four per call.
    • Open questions, or ones that need a paragraph of context: number them in text and wait for the answers.
  4. Challenge, don't just collect. Every round, name the assumption the idea leans on hardest and test it: what breaks if it is false, what the cheaper version is, whether it is needed at all. Push harder when the user asks for it.
  5. Recompute after each round. New answers unlock new questions and retire others. Then run the Spec-Probe: could you write every Spec section now, with no section resting on a guess? If not, ask the next round. If yes, offer one more round or the Spec.
  6. Hold the scope line. Phase cut, ordering and approach belong to /crew:plan: note each such idea for Notes for planning and move on.
  7. Close by restating the Spec and getting the user's confirmation.

The roast is done when the Spec-Probe passes and the user confirms; until then, the next round is the next step.

Outside /crew:brief

Works without .planning/: run the same rounds and give the clarified result back in the conversation. Write files only when asked; to capture it as crew state, point to /crew:brief.

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.