Agent skill
speci5.brainstorm
Elaborate and expand user input into structured idea documents in .spec/ideas/. Takes rough thoughts, researches patterns, surfaces trade-offs and open questions. WHEN: /brainstorm, brainstorm idea, new idea, capture idea, explore concept, elaborate on, think about, research idea. DO NOT USE FOR: writing specs (use specify), creating plans (use plan), checking implementation (use check).
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/speci5-brainstorm
SKILL.md
Brainstorm
Turn rough ideas into well-structured idea documents through collaborative dialogue.
Start by understanding the project context and the user's intent, then ask clarifying questions before elaborating. The goal is a thinking artifact — not a formal spec.
Triggers
Activate this skill when the user wants to:
- Capture a new idea or concept
- Elaborate on a rough thought
- Research patterns, prior art, or trade-offs for an idea
- Update or refine an existing idea document
Rules
- Use kebab-case for all filenames (e.g.,
user-auth.md). - Do NOT create feature or story files — that's the specify skill's job.
- If updating an existing idea, preserve human-written content and append/refine rather than overwrite.
- Keep language clear and concise — this is a thinking artifact, not a formal spec.
- YAGNI ruthlessly — Strip unnecessary complexity. Simpler ideas are better ideas. If a feature isn't clearly needed, leave it out.
Conversation Style
- One question at a time. Don't overwhelm the user with multiple questions in one message.
- Prefer multiple choice when possible — easier to answer than open-ended.
- Be flexible — go back and clarify when something doesn't make sense.
Steps
- Explore project context — Scan
.spec/ideas/for existing ideas, but also check the codebase, docs, and recent changes to understand what's already built and what patterns are in use. Ideas don't exist in a vacuum. - Check scope — If the user's input describes multiple independent subsystems or a very large surface area, flag this immediately. Help decompose into smaller, independent ideas before elaborating. Each idea should be a single cohesive concept.
- Ask clarifying questions — Before writing anything, ask questions one at a time to understand purpose, constraints, and success criteria. Don't skip this even for "simple" ideas — unexamined assumptions cause the most wasted work.
- Propose 2-3 approaches — Present different ways to tackle the idea, with trade-offs for each. Lead with your recommendation and explain why.
- Write or update — Once you and the user are aligned, write the idea document. If the input closely relates to an existing idea file, update it. Otherwise, create a new file in
.spec/ideas/. - Present — Show the user the resulting idea document and highlight any open questions.
Output Format
See output template.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?