agentleFS
PaulRBergPaulRBerg/dot-agents/skills/brainstorm-with-me/SKILL.md

brainstorm-with-me

Skill6 starsChanged 7 days agoMITView on GitHub

This skill is coordination-exempt: skip the ai-coord gate for its declared work. Develop a non-obvious, useful concept through a focused creative interview. Then return one substantial synthesis. 1. Extract the objective, audience, constraints, taste signals, and what feels stale about the obvious answers. Investigate facts available from the conversation, codebase, or supplied evidence instead of asking for them. 2. Ask exactly one concrete question per turn. Format every question turn as ### 💡 Question <N> — <topic>, followed by **🧭…

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    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. Very specific:

    Ask exactly one concrete question per turn.

    line 19
  2. Distinctive:

    Treat three to five questions as a soft target.

    line 26
  3. Distinctive:

    Do not end the completed brainstorm with another question.

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Brainstorm With Me

This skill is coordination-exempt: skip the ai-coord gate for its declared work.

Develop a non-obvious, useful concept through a focused creative interview. Then return one substantial synthesis.

Workflow

  1. Extract the objective, audience, constraints, taste signals, and what feels stale about the obvious answers. Investigate facts available from the conversation, codebase, or supplied evidence instead of asking for them.
  2. Ask exactly one concrete question per turn. Format every question turn as ### 💡 Question <N> — <topic>, followed by **🧭 Context**, optional **🎯 Answer shape**, and **❓ Question**, with each label on its own line. Keep the context brief and put exactly one question in the final section. When they make an abstract choice easier to answer, use two to four concise examples or response anchors. Do not turn them into an idea list.
  3. Choose the question whose answer would most change the creative direction. Favor specific preferences, tensions, exclusions, emotional or practical effects, acceptable tradeoffs, and reactions to concrete contrasts. Avoid broad prompts that ask the user to do the ideation.
  4. Treat three to five questions as a soft target. Stop earlier when the prompt and answers already supply a strong creative brief. Continue only while another answer could materially improve the final recommendation. When useful, briefly acknowledge an answer as ✅ Noted: <preference or implication>. Do not show progress recaps, candidate concepts, or interim synthesis.
  5. Explore possibilities internally as the interview develops. Change lenses selectively through analogy, assumption reversal, an unrelated domain, another stakeholder, subtraction, combination, changed scale or time horizon, extreme constraints, or the useful fragment of a deliberately bad idea. Track promising fragments and rejected patterns. Preserve distinct directions. Before choosing a direction, compare them against the user's evidence.
  6. Once further questions would add little value, return the first full synthesis of the session. Make it materially fuller than the question turns. Make it concrete rather than exhaustive. Use this structure:
    • ### 🧭 Creative brief: objective, audience, constraints, and taste signals.
    • ### 💡 Recommended concept: the concept, how it works, and the experience it creates.
    • ### ✨ Why it could work: fit to the interview evidence and meaningful difference from obvious alternatives.
    • Optional ### 🔀 Strong alternatives: at most two genuinely different directions and their tradeoffs.
    • Optional ### ⚠️ Risks and open choices: only uncertainties that could change execution.
    • ### 🧪 Smallest test: the minimal action and observable success signal.
  7. Do not end the completed brainstorm with another question. If the user explicitly asks to keep diverging or to stop the interview early, honor that request while preserving the same final-report shape when synthesis is requested.

Completion requires a chosen or best-supported concept whose usefulness and distinctiveness are clear, plus a smallest test with an observable success signal grounded in the interview.

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.