Brainstorm
🥷 AI agents for knowledge work, not code
npx -y skills add qkitzero/kage-bunshin --skill brainstormAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Multi-perspective brainstorming session. Use when the user wants to generate and evaluate ideas on a topic. Triggered by idea generation, brainstorming, "any good ideas?", wanting new solutions, or creative suggestions.
SKILL.md
3.2 KB, as published. Nobody here has run it
Brainstorm Skill
A workflow that creates an Agent Team of Ideator and Analyst to perform iterative idea divergence and convergence through multi-round discussion.
Workflow
Step 1: Understand the Theme
Accurately understand the theme, challenge, or problem presented by the user. If ambiguous, confirm the following before proceeding:
- Target domain (business, technology, creative, etc.)
- Constraints (budget, timeline, tech stack, etc.)
- Required granularity of deliverables
Step 2: Retrieve Past Notebook Context
If NOTEBOOK_PATH is set, search for past entries related to the user's theme:
- Grep
$NOTEBOOK_PATH/ideas/for keywords from the theme (searching title, tags, and body) - Read up to 3 matching entries (frontmatter + first 200 characters of body)
- Include the retrieved content as context when creating the team
Skip this step if zero matches or NOTEBOOK_PATH is not set.
Step 3: Create Agent Team
Create a team with the following teammates using their agent definitions from .claude/agents/:
Teammates:
ideator— generates creative ideasanalyst— evaluates and challenges ideas
Team instructions:
Theme: [User's theme]
[Past Notebook context if found]
Collaboration protocol:
1. Ideator generates 5-7 ideas with summary, rationale, and impact assessment
2. Analyst evaluates each idea (Feasibility, Impact, Novelty, Risk on 1-5 scale) and provides improvement suggestions
3. Ideator refines weak ideas based on Analyst's feedback, may propose new ideas inspired by the critique
4. Analyst re-evaluates the refined ideas
5. Continue iterating until both agree on a final ranked list (aim for 2-3 rounds)
Rules:
- Output language: Use the language specified by OUTPUT_LANGUAGE env var. If not set, match the user's language (default: English)
- If NOTEBOOK_PATH is set: Ideator saves final ideas to $NOTEBOOK_PATH/ideas/, Analyst saves evaluation to $NOTEBOOK_PATH/reviews/
- Each agent writes its own deliverables with structured frontmatter
Step 4: Synthesis & Ranking
After the team completes its discussion, create the final output:
## Brainstorming Results: [Theme]
### Discussion Summary
[Brief description of how ideas evolved through iteration]
### #1: [Idea Name]
- **Overall Score**: X.X / 5.0
- **Summary**: ...
- **Scores**: Feasibility X | Impact X | Novelty X | Risk X
- **Strengths**: ...
- **How it was refined**: ...
### #2: [Idea Name]
...
### #3: [Idea Name]
...
### Other Ideas
- [Idea Name]: Overall Score X.X - [One-line summary]
### Next Steps
- Recommend `/evaluate` for detailed assessment of promising ideas
- Recommend `/plan-project` for turning ideas into projects
Output Format
Present the following to the user:
- Brief summary of the team discussion and iteration process
- Top 3 ideas in detail (with scores, evaluation, and how they were refined)
- List of remaining ideas
- Next step suggestions
- Notebook save notification if applicable