agentleFS
darrenCWJdarrenCWJ/agent-skills-scaffold/.claude/skills/brainstorm/SKILL.md

brainstorm

Skill0 starsChanged 3 months agoMITView on GitHub

Structured divergent-then-convergent ideation before committing to an approach. Use whenever the solution space is open — the user says brainstorm, ideas, options, what are the ways, how should we, name this, or design this — and BEFORE writing any ADR that would otherwise have only one alternative. If the first idea is about to become the plan without competition, this skill is overdue.

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use playtest-report, team-polish 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.
  • CoralGameSkill

    playtest-report

    Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.

    Also in 9 projects that use this

  • CoralGameSkill

    team-polish

    Orchestrate the polish team: coordinates performance-analyst, technical-artist, sound-designer, and qa-tester to optimize, polish, and harden a feature or area for release quality.

    Also in 11 projects that use this

  • lilangMaxSkill

    soak-test

    Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.

    Also in 6 projects that use this

  • lilangMaxSkill

    team-live-ops

    Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.

    Also in 6 projects that use this

  • +8 more from CoralGame

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 158 edited versions we read keep these.

  • 1

    Distinctive

    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:

    Keep the top 2–3 as genuine finalists with trade-offs stated.

    line 28

“Common” is measured: kept by at least 6 in 10 of the 158 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

Brainstorm

The first idea is rarely the best idea; it is merely the most available one. Brainstorming exists to force real alternatives into existence before convergence — and to converge honestly afterward. Never evaluate while generating; never generate while evaluating.

Phase 1 — Frame (2 minutes)

State: the actual problem (not the assumed solution), hard constraints (real ones — budget, compatibility, deadline), and what "good" means here. A mis-framed brainstorm generates 20 answers to the wrong question. Challenge the frame once: "is this the real problem, or a symptom of a bigger one?"

Phase 2 — Diverge

Generate 8–15 candidates. Quantity IS the mechanism — the obvious ideas must be exhausted before the interesting ones appear. Rules:

  • No evaluation during generation. No "but that won't work because..." — capture it and move on. Killing ideas early kills the branches that would have grown from them.
  • Force variety with lenses. Generate at least one idea from each: the conventional answer; the opposite/inversion of the conventional answer; what you'd do with 10x the resources; what you'd do with 1/10th; how an adjacent domain solves this (analogy); removing the assumed-fixed constraint; combining two earlier ideas; and do nothing (always a candidate — its cost is the baseline every idea must beat).
  • Include one idea you dislike. Authored aversion is data about you, not the idea.
  • For substantial problems, fan out 2–3 subagents with different lenses (fan-out skill) — independent generation beats one mind's clustering.

Phase 3 — Converge (criteria first, then scores)

  1. Write the criteria BEFORE looking back at the list (impact, effort, risk, reversibility, fit with standing ADRs — pick 3–5, weight them). Criteria written after seeing options become post-hoc rationalization of the favorite.
  2. Cluster the candidates; name the clusters — clusters often reveal a dimension of choice nobody stated.
  3. Score the plausible ones against the criteria. Kill criteria first (violates a hard constraint = out, no matter how clever).
  4. Keep the top 2–3 as genuine finalists with trade-offs stated. Present to the user when the choice is theirs; decide when it's yours.

Phase 4 — Record

The chosen approach AND the rejected finalists (with reasons) go into an ADR (design-decisions skill). The brainstorm's raw list goes into the plan file or a docs/plans/<slug>-brainstorm.md appendix — discarded ideas are free inventory for the next problem.

Anti-patterns

  • Brainstorming to decorate a decision already made — dishonest and slower than just deciding.
  • Converging after 3 ideas because one "feels right" (availability bias wearing a process costume).
  • Letting the loudest constraint (usually effort) silently become the only criterion.

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.