AgentSkillsCN

rsn-creating-ideas

生成新颖的创意与富有创造力的问题解决方案。运用横向思维、SCAMPER法则、第一性原理以及各类创意发想框架,打破传统思维定式。对创意进行原创性、可行性和创意品质的多重审核,帮助用户突破创意瓶颈,强化薄弱概念。当您需要头脑风暴、卡壳于某个难题、亟需新鲜视角,或希望打破常规思维时,可选用此技能。触发关键词包括:“头脑风暴”、“生成创意”、“创造性思考”、“卡壳了”、“换个新视角”。

SKILL.md
--- frontmatter
name: rsn-creating-ideas
description: Generates novel ideas and creative solutions to problems. Applies lateral thinking, SCAMPER, first principles, and ideation frameworks to break conventional patterns. Audits ideas for originality, feasibility, and creative quality. Fixes creative blocks and strengthens weak concepts. Use when brainstorming, stuck on a problem, need fresh perspectives, or breaking conventional thinking. Triggers on "brainstorm", "generate ideas", "think creatively", "stuck", "fresh perspective".
license: Complete terms in LICENSE.txt

Creating Ideas

Expert knowledge of ideation methodologies, lateral thinking, and creative problem-solving. Grounded in de Bono (lateral thinking, Six Hats), Osborn (brainstorming, SCAMPER), design thinking, and first principles reasoning.

Mode Detection

User SaysMode
"brainstorm", "generate ideas", "think creatively", "fresh perspective"APPLY
"is this idea good", "evaluate concept", "how original is this"AUDIT
"stuck", "no ideas", "concept isn't working", "make this more creative"FIX

If ambiguous: "APPLY creative frameworks, AUDIT existing ideas, or FIX creative blocks?"

Relationship to rsn-reasoning-problems

This skill generates ideas. For reasoning about ideas once generated:

  • Use rsn-reasoning-problems.dialectical to evaluate trade-offs
  • Use rsn-reasoning-problems.analogical for deeper domain transfer analysis
  • Use rsn-reasoning-problems.causal to plan execution

APPLY Mode

Core Principles

PrincipleOne-Liner
Diverge then ConvergeGenerate many ideas first, evaluate later
Suspend Judgment"Yes, and..." not "No, but..." during ideation
Quantity Breeds QualityMore ideas → higher chance of breakthrough
Combine and BuildBest ideas often merge multiple concepts
Challenge AssumptionsQuestion every "obvious" constraint

Top 10 Techniques

  1. First Principles: Strip to fundamentals, rebuild from truth
  2. SCAMPER: Substitute, Combine, Adapt, Modify, Put to other uses, Eliminate, Reverse
  3. Lateral Thinking: Escape dominant patterns via provocation
  4. Analogical Transfer: Borrow solutions from other domains
  5. Constraint Manipulation: Remove, add, or flip constraints
  6. Random Stimulation: Force connections with unrelated inputs
  7. Reverse Brainstorming: "How could we make this worse?"
  8. Worst Possible Idea: Start terrible, extract useful elements
  9. Six Thinking Hats: Structured parallel thinking perspectives
  10. Morphological Analysis: Systematic combination of attributes

Full catalog: references/patterns.md

Process

  1. Define challenge — What problem are we solving? What does success look like?
  2. Gather inputs — Current constraints, past attempts, domain knowledge
  3. Select technique(s) — Match technique to problem type
  4. Diverge — Generate 10-50+ ideas without judgment
  5. Incubate — Allow unconscious processing if time permits
  6. Converge — Cluster, combine, evaluate, select
  7. Develop — Strengthen selected ideas

Output Format

markdown
## Creative Exploration: [Challenge]

### Challenge Reframe
- Original: [How it was stated]
- Reframed: [More generative framing]

### Technique Applied: [Name]

### Ideas Generated
1. **[Idea Name]**: [One-line description]
   - Mechanism: [How it works]
   - Novel element: [What's new]

[Repeat for top ideas]

### Combinations Worth Exploring
- [Idea A] + [Idea B] → [Combined concept]

### Recommended Next Steps
1. [Most promising direction]

AUDIT Mode

Evaluation Dimensions

DimensionCheck
OriginalityIs this genuinely new or recombined familiar?
FeasibilityCan this actually be built/implemented?
ValueDoes this solve a real problem meaningfully?
ClarityIs the concept clear and communicable?
DefensibilityCan competitors easily copy this?
ScalabilityDoes this grow or stay niche?

Full rubric with 0-3 criteria: references/audit-rubric.md

Process

  1. Understand the idea — Can you explain it simply?
  2. Score dimensions — 0-3 on each criterion
  3. Identify strengths — What's working?
  4. Identify gaps — What's missing or weak?
  5. Suggest improvements — How to strengthen?

Output Format

markdown
## Idea Audit: [Concept Name]

**Score:** X/18 | **Verdict:** [Breakthrough/Promising/Incremental/Weak]

### Dimension Scores
| Dimension | Score | Assessment |
|-----------|-------|------------|
| Originality | /3 | [Finding] |
| Feasibility | /3 | [Finding] |
| Value | /3 | [Finding] |
| Clarity | /3 | [Finding] |
| Defensibility | /3 | [Finding] |
| Scalability | /3 | [Finding] |

### Strengths
- [What's working]

### Gaps
- [What's missing]

### Enhancement Recommendations
1. [Specific improvement]

FIX Mode

Common Creative Blocks

BlockSymptomFix Technique
Functional FixednessCan only see obvious usesAnalogical transfer, SCAMPER
Einstellung EffectStuck on first solutionConstraint removal, reverse brainstorm
Analysis ParalysisOverthinking, no outputWorst idea first, time pressure
Premature JudgmentKilling ideas too earlyDiverge/converge separation
Domain BlindnessOnly seeing industry normsRandom stimulation, cross-domain
Scope CreepIdea too complexFirst principles, constraint addition

Diagnostic Process

  1. Identify block type — What's preventing progress?
  2. Select antidote technique — Match technique to block
  3. Apply technique — Generate new options
  4. Extract value — What's useful in the output?
  5. Iterate — Refine or try another technique

Idea Strengthening Process

  1. Isolate weakness — What specifically is weak?
  2. Diagnose cause — Why is it weak?
  3. Apply targeted fix — Specific enhancement
  4. Validate improvement — Re-audit the idea

Output Format

markdown
## Creative Fix: [Problem/Block]

### Diagnosis
- **Block type:** [Category]
- **Symptom:** [What's happening]
- **Root cause:** [Why it's happening]

### Technique Applied: [Name]

### Before
[Original state/idea]

### After
[Improved state/ideas]

### What Changed
- [Specific improvement and why it helps]

Examples: references/examples.md


Technique Selection Guide

Problem TypeBest Techniques
Need more ideasSCAMPER, Random Stimulation, Worst Idea
Stuck on one solutionConstraint Manipulation, Reverse Brainstorm
Need breakthroughFirst Principles, Analogical Transfer
Too many ideasSix Hats evaluation, Morphological narrowing
Idea too vagueFirst Principles, Constraint Addition
Idea too complexElimination, Core extraction
Need team alignmentSix Thinking Hats, structured brainstorm

Failure Handling

SituationAction
Technique produces nothing usefulSwitch technique; try opposite approach
All ideas seem badUse "Worst Idea" to lower pressure; extract elements
Can't escape existing solutionAdd extreme constraint; remove core assumption
Ideas too incrementalAsk "What would 10x require?"; analogize from distant domain
Overwhelmed by optionsApply Six Hats structure; force ranking
Time pressureUse rapid SCAMPER (30 seconds per letter)

Anti-patterns: references/anti-patterns.md


Boundaries

In scope:

  • Ideation and brainstorming
  • Problem reframing
  • Creative problem-solving
  • Concept development
  • Breaking mental blocks
  • Innovation methodologies

Out of scope:

  • Implementation planning (use rsn-reasoning-problems.causal)
  • Market validation (use rsn-perceiving-information)
  • Visual/artistic design
  • Technical feasibility deep-dive

References

FileContent
patterns.mdFull technique catalog
audit-rubric.mdDetailed scoring criteria
anti-patterns.mdWhat to avoid
examples.mdWorked examples