| name | brainstorm |
| description | Structured ideation with constraint mapping, divergent exploration, and convergent selection |
| layer | hub |
| category | workflow |
| triggers | ["/brainstorm","brainstorm ideas","what are the options","how could we approach this","explore alternatives","think of ways to"] |
| inputs | [{"challenge":"The problem, question, or goal to brainstorm about"},{"constraints":"Hard constraints that solutions must satisfy (optional)"},{"preferences":"Soft preferences to guide selection (optional)"},{"quantity":"Number of ideas to generate (optional, default 5-7)"}] |
| outputs | [{"constraintMap":"Explicit map of hard constraints, soft preferences, and degrees of freedom"},{"ideas":"Ranked list of ideas with descriptions, tradeoffs, and feasibility notes"},{"recommendation":"Top 1-3 recommended approaches with reasoning"},{"nextSteps":"Concrete actions to move from ideation to execution"}] |
| linksTo | ["plan","research","scout"] |
| linkedFrom | ["plan","cook","team"] |
| preferredNextSkills | ["plan","research","scout"] |
| fallbackSkills | ["research"] |
| riskLevel | low |
| memoryReadPolicy | selective |
| memoryWritePolicy | selective |
| sideEffects | ["Produces ideation document in working memory","May invoke research for domain knowledge"] |
Brainstorm Skill
Purpose
Generate diverse, well-structured ideas for solving a problem or approaching a challenge. This skill separates divergent thinking (generating options) from convergent thinking (selecting the best ones) to prevent premature optimization of the solution space.
The goal is not to find THE answer. The goal is to find the right set of options and make the tradeoffs visible so the best decision can be made.
Workflow
Phase 1: Problem Framing
-
Restate the challenge in precise terms. Often the way a problem is framed determines what solutions are visible.
-
Build the constraint map:
- Hard constraints: Non-negotiable requirements. Solutions that violate these are automatically disqualified.
- Soft preferences: Desired but flexible. Solutions that satisfy these are preferred but not required.
- Degrees of freedom: Dimensions where we have choices. These are the axes of exploration.
- Assumptions: Things we are taking as given. Worth questioning -- sometimes relaxing an assumption opens new options.
-
Identify the decision type:
- Reversible (two-way door): Bias toward action. Pick something and iterate.
- Irreversible (one-way door): Invest in thorough analysis. These decisions are expensive to undo.
Phase 2: Divergent Exploration
-
Generate ideas using multiple lenses:
Lens 1: Direct solutions -- What is the most straightforward way to solve this?
Lens 2: Analogies -- How has this type of problem been solved in other domains?
Lens 3: Inversion -- Instead of solving the problem, what if we removed the conditions that create it?
Lens 4: Decomposition -- Can the problem be broken into subproblems that are each easier to solve?
Lens 5: Extreme positions -- What would the solution look like if we optimized for only ONE dimension (speed, simplicity, flexibility, cost)?
Lens 6: Existing solutions -- What off-the-shelf tools, libraries, or patterns already solve this?
Lens 7: Future-backward -- If this were already solved, what would the solution look like? Work backward from there.
-
For each idea, capture:
