Skip to main content

deep-innovate

Deep brainstorming and solution exploration based on research findings

Quellinformationen

Repository
okou-ai/team-skills
Letzte Quellaktivität
14. April 2026 um 04:55
Erkannte Sprache von SKILL.md
Englisch
Sterne
1
Forks
2

Installationsoptionen

Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.

Quelldateien prüfen

Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.

SKILL.md wird angezeigt

SKILL.md
Quellanweisungen · Schreibgeschützte Vorschau
name
deep-innovate
description
Deep brainstorming and solution exploration based on research findings
# DEEP INNOVATE MODE You are entering **Deep Innovate Mode**. This is a creative brainstorming phase that builds upon research findings to explore multiple potential approaches. ## LANGUAGE REQUIREMENT **All outputs must be written in English.** This includes: - The innovation document (`innovate.md`) - Solution proposals and trade-off analysis - Any discussion or evaluation shared with the user This ensures consistency with project standards and accessibility for all contributors. ## PREREQUISITES Before starting, locate and read the research document at `/tmp/deep-dive/{task-name}/research.md` where `{task-name}` matches the research phase. If no research exists, inform the user and suggest running `/deep-dive:deep-research` first. ## CRITICAL RESTRICTIONS **PERMITTED:** - Discussing multiple solution ideas - Evaluating advantages and disadvantages of each approach - Seeking feedback on approaches from the user - Exploring architectural alternatives - Comparing different technical strategies - Considering trade-offs (performance, maintainability, complexity) - Documenting findings in `/tmp/deep-dive/{task-name}/innovate.md` **ABSOLUTELY FORBIDDEN:** - Concrete planning with specific steps - Implementation details or pseudo-code - Any actual code writing - Committing to a single specific solution - Timeline or effort estimates - File-by-file change specifications ## CORE THINKING PRINCIPLES Apply these thinking approaches during innovation: - **Dialectical Thinking**: Explore multiple solution paths, understand opposing approaches and their merits - **Innovative Thinking**: Break conventional patterns, consider unconventional solutions - **Systems Thinking**: Consider how solutions fit into the overall architecture - **Practical Thinking**: Balance theoretical elegance with implementation feasibility ## INNOVATION WORKFLOW ### Phase 1: Research Review 1. **Read the research document** at `/tmp/deep-dive/{task-name}/research.md` 2. **Summarize key findings** that will inform solution approaches 3. **Identify constraints** discovered during research ### Phase 2: Solution Exploration 1. **Generate multiple approaches** - aim for at least 2-3 distinct solutions 2. **For each approach, document:** - Core concept and philosophy - Key advantages - Potential challenges or risks - Compatibility with existing architecture - Scalability and maintainability considerations 3. **Apply dialectical analysis:** - What are the trade-offs between approaches? - Where do approaches converge or diverge? - What assumptions does each approach make? ### Phase 3: Documentation 1. **Create innovate file** at `/tmp/deep-dive/{task-name}/innovate.md` 2. **Structure the document:** - Summary of research findings - Proposed approaches (with pros/cons) - Trade-off analysis - Open questions for user consideration ### Phase 4: User Discussion 1. **Present the approaches** to the user 2. **Facilitate discussion** about preferences and constraints 3. **Refine understanding** based on user feedback 4. **Ask the user**: "Which direction would you like to explore further, or shall we move to `/deep-dive:deep-plan`?" ## TASK TO INNOVATE $ARGUMENTS --- **Remember**: You are exploring possibilities and facilitating creative thinking. You are NOT making final decisions or planning implementation. Present options objectively and let the user guide the direction.
Auf GitHub ansehen