agentleFS
zhouguoqingzhouguoqing/QianYuan.AIAgenticFramework/.agents/skills/brainstorm/SKILL.md

brainstorm

Skill35 starsChanged 6 months agoApache-2.0View on GitHub

Collaborative discovery and design framing for ambiguous or high-risk work. Use when requirements are unclear, multiple approaches are possible, or you need to turn an idea into a validated design brief before planning or coding.

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

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. 0 are common, 16 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 157 edited versions we read keep these.

  • 16

    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 skill to convert rough ideas into clear, reviewable design outputs through structured dialogue.

    line 12
  2. Very specific:

    If docs/SUMMARY.md exists, read it first.

    line 26
  3. Distinctive:

    If docs conflict with code or user intent, use the available input/question tool before broad changes.

    line 29
  4. Distinctive:

    Keep this pass focused. Only gather what is needed for the current idea.

    line 33
  5. Distinctive:

    Ask targeted questions sequentially to remove ambiguity with input/question tool:

    line 37
  6. Distinctive:

    Do not jump to implementation details too early.

    line 46
  7. Distinctive:

    After presenting all approaches, use input/question tool to let the user pick their preferred approach.

    line 62
  8. Distinctive:

    After presenting each phase, use input/question tool immediately to ask whether to:

    line 79
  9. Distinctive:

    After you and the user have worked through requirements and the design is validated, determine the next actions.

    line 87
  10. Distinctive:

    Use input/question tool to present the user with three high-level next actions:

    line 89
  11. Distinctive:

    If the user picks Write artifacts, proceed to Step 6.

    line 95
  12. Very specific:

    If the user picked Write plan immediately, initiate a handoff to use skill write-plan using the current brainstorming context.

    line 100
  13. Distinctive:

    If the user picked End session, simply stop.

    line 102
  14. Distinctive:

    Only perform this step after the user has explicitly chosen "Write artifacts" during Step 5.

    line 106
  15. Distinctive:

    Do not write production code or make implementation changes in this skill in the brainstorm session.

    line 121
  16. Distinctive:

    Keep interaction lightweight and iterative.

    line 122
  17. Distinctive:

    Prefer clarity over completeness when uncertain.

    line 123
  18. Distinctive:

    Keep assumptions explicit.

    line 125

“Common” is measured: kept by at least 6 in 10 of the 157 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

Overview

Use this skill to convert rough ideas into clear, reviewable design outputs through structured dialogue.

The goal is to:

  1. Clarify requirements and constraints
  2. Explore alternatives with trade-offs
  3. Produce a concrete, validated design brief in artifacts or a handoff to planning

Workflow

Step 1: Gather Project Context

Load only the project context relevant to the current idea:

  • If docs/SUMMARY.md exists, read it first.
  • Load only task-relevant detail docs.
  • Prioritize Code Standard docs for implementation conventions.
  • If docs conflict with code or user intent, use the available input/question tool before broad changes.

Also check key implementation files relevant to the idea and note constraints from existing architecture, dependencies, and conventions.

Keep this pass focused. Only gather what is needed for the current idea.

Step 2: Clarify Requirements

Ask targeted questions sequentially to remove ambiguity with input/question tool:

  • Focus on:
    • Objective and user value
    • Scope boundaries
    • Constraints (technical, UX, performance, timeline)
    • Success criteria
    • Non-goals

Do not jump to implementation details too early.

Step 3: Explore Approaches

Propose 2-5 viable approaches.

For each approach, include:

  • Short summary
  • Pros
  • Cons / risks
  • Complexity estimate
  • Recommended use conditions

Lead with your recommended option and explain why it best fits the project context and constraints.

After presenting all approaches, use input/question tool to let the user pick their preferred approach. List the summary options. Example:

  1. Approach A, short summary
  2. Approach B, short summary
  3. Approach C, short summary
  4. Other (please specify)

with the tag added for the recommended approach to guide the user.

Step 4: Present the Design Incrementally

Once requirements are clear, present the design incrementally in logical phases (about 200-300 words per phase) to avoid overwhelming the user.

  • Phase 1: Foundation - Problem framing, goals, and proposed architecture/flow.
  • Phase 2: Technical Details - Data model, interfaces, error handling, and edge cases.
  • Phase 3: Delivery - Testing/verification strategy and rollout considerations (if applicable).

After presenting each phase, use input/question tool immediately to ask whether to:

  1. Proceed to the next phase
  2. Adjust the current phase
  3. Revisit a previous phase

Step 5: Close the Loop

After you and the user have worked through requirements and the design is validated, determine the next actions.

  1. Use input/question tool to present the user with three high-level next actions:
    • "Write plan immediately (in current context)" - skip the artifact step and move straight to a write-plan handoff.
    • "Write artifacts" - continue by authoring the brainstorm documents described in Step 6.
    • "Implement immediately (skip design artifacts and planning)" - if the design is very clear and low-risk, the user may choose to skip both artifacts and planning and move straight to implementation.
    • "End session (already provided enough information for user)" - stop; the conversation has produced enough insight for now.
  2. If the user picks Write artifacts, proceed to Step 6. Once the draft artifacts exist, use input/question tool again to validate them with options:
    • "Write plan with current artifacts, context"
    • "End session - artifacts are sufficient for now"
    • "Need changes" (free-form text) - collect the feedback, revise the artifacts, and re-ask.
  3. If the user picked Write plan immediately, initiate a handoff to use skill write-plan using the current brainstorming context; no additional artifact validation is required.
  4. If the user picked End session, simply stop. The information collected so far is considered sufficient.

Step 6 (optional): Write Brainstorm Artifacts

Only perform this step after the user has explicitly chosen "Write artifacts" during Step 5.

Persist results to the standardized location:

  • Directory: docs/.brainstorms/YYMMDD-HHmm-<topic-slug>/
  • Main file (required): docs/.brainstorms/YYMMDD-HHmm-<topic-slug>/SUMMARY.md
  • Optional supporting files:
    • docs/.brainstorms/YYMMDD-HHmm-<topic-slug>/section-01-<slug>.md
    • docs/.brainstorms/YYMMDD-HHmm-<topic-slug>/section-02-<slug>.md
    • etc.

SUMMARY.md format: strictly follow the template inside references/summary-template.md.

Rules

  • Do not write production code or make implementation changes in this skill in the brainstorm session.
  • Keep interaction lightweight and iterative; every step should be run in the same session.
  • Prefer clarity over completeness when uncertain; ask a follow-up question.
  • Align all recommendations with project documentation and standards.
  • Keep assumptions explicit; do not guess silently.
  • Think before coding: Surface assumptions explicitly. If multiple interpretations exist, present them — don't pick silently. If a simpler approach exists, say so and push back when warranted.

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.