Free SKILL.md scraped from GitHub. Clone the repo or copy the file directly into your Claude Code skills directory.
npx versuz@latest install onfire7777-universal-ai-skills-library-plugin-codex-skills-creative-connections-enginegit clone https://github.com/onfire7777/universal-ai-skills-library.gitcp universal-ai-skills-library/SKILL.MD ~/.claude/skills/onfire7777-universal-ai-skills-library-plugin-codex-skills-creative-connections-engine/SKILL.md--- name: creative-connections-engine description: A meta-skill for radical creativity, lateral thinking, idea synthesis, conceptual blending, bisociation, mind mapping, and finding unexpected connections between disparate ideas. Use when brainstorming, generating novel ideas, synthesizing across domains, creating mind maps, finding analogies, breaking creative blocks, or when the user asks for creative thinking, lateral thinking, divergent thinking, or unique perspectives. Combines the best frameworks from Arthur Koestler's bisociation theory, Edward de Bono's lateral thinking, Fauconnier & Turner's conceptual blending, TRIZ, SCAMPER, synectics, random stimulation, forced connections, and mind mapping methodologies. license: Unspecified --- # Creative Connections Engine ## Purpose This skill transforms the agent into a radically creative thinker that sees connections others miss. It synthesizes ideas across domains, generates novel combinations, and produces genuinely original thinking rather than conventional responses. ## Core Philosophy > "The creative act is not an act of creation in the sense of the Old Testament. It does not create something out of nothing; it uncovers, selects, re-shuffles, combines, synthesizes already existing facts, ideas, faculties, skills." — Arthur Koestler, *The Act of Creation* Creativity is **combinatorial**. Every breakthrough comes from connecting ideas that were previously unconnected. This skill provides systematic methods for finding those connections. ## When to Use - Brainstorming sessions or ideation - Finding novel solutions to stuck problems - Synthesizing ideas from multiple domains - Creating mind maps of interconnected concepts - Breaking out of conventional thinking patterns - Generating analogies and metaphors - Cross-pollinating ideas between fields - Reframing problems from unexpected angles ## The Seven Engines ### Engine 1: Bisociation (Koestler) The most powerful creativity technique. Bisociation means connecting two previously unrelated **matrices of thought** (frames of reference, contexts, or associative fields). **Process:** 1. **Identify the problem domain** (Matrix A) 2. **Select a completely unrelated domain** (Matrix B) — the more distant, the better 3. **List the core principles, patterns, and structures** of Matrix B 4. **Force connections** between Matrix B's patterns and Matrix A's problem 5. **Evaluate** which connections spark genuine insight **Example domains to bisociate with:** - Biology (evolution, ecosystems, symbiosis, immune systems) - Music (harmony, counterpoint, rhythm, improvisation) - Architecture (load-bearing, negative space, cantilevers) - Cooking (fermentation, reduction, emulsification, layering) - Sports (strategy, momentum, positioning, teamwork) - Nature (fractals, emergence, migration, metamorphosis) - Games (game theory, rules, strategy, bluffing) - History (revolutions, trade routes, cultural exchange) ### Engine 2: Lateral Thinking (de Bono) Break free from vertical (logical, step-by-step) thinking by deliberately disrupting patterns. **Six Thinking Hats (apply all six to any problem):** - **White Hat**: Pure facts and data — what do we actually know? - **Red Hat**: Emotions and intuition — what does your gut say? - **Black Hat**: Critical judgment — what could go wrong? - **Yellow Hat**: Optimism — what's the best possible outcome? - **Green Hat**: Creativity — what wild alternatives exist? - **Blue Hat**: Process — how should we think about this? **Lateral Thinking Provocations:** - **Reversal**: What if the opposite were true? - **Exaggeration**: What if we 10x or 1/10x a key variable? - **Distortion**: What if we change the sequence, timing, or relationships? - **Wishful thinking**: What if there were no constraints at all? - **Random entry**: Pick a random word/image and force a connection ### Engine 3: Conceptual Blending (Fauconnier & Turner) Systematically blend two input mental spaces into a novel "blended space." **Process:** 1. **Input Space 1**: Define the first concept with its structure, roles, and relationships 2. **Input Space 2**: Define the second concept with its structure, roles, and relationships 3. **Generic Space**: Identify the abstract structure shared by both 4. **Cross-Space Mapping**: Map elements from Space 1 to Space 2 5. **Blended Space**: Create the novel combination by selectively projecting elements from both inputs 6. **Emergent Structure**: Identify what's NEW in the blend that wasn't in either input ### Engine 4: SCAMPER Systematic creative modification of existing ideas: - **S**ubstitute: What can be replaced? - **C**ombine: What can be merged? - **A**dapt: What can be borrowed from elsewhere? - **M**odify/Magnify/Minimize: What can be changed in scale or form? - **P**ut to other uses: What else could this be used for? - **E**liminate: What can be removed? - **R**everse/Rearrange: What can be flipped or reordered? ### Engine 5: Synectics (Making the Strange Familiar, Making the Familiar Strange) **Four types of analogy:** 1. **Direct Analogy**: Find a parallel in nature or another field 2. **Personal Analogy**: Become the problem — "If I were the database, what would I feel?" 3. **Symbolic Analogy**: Compress the problem into a metaphor or image 4. **Fantasy Analogy**: What would a magic solution look like? ### Engine 6: Random Stimulation & Forced Connections **Process:** 1. Pick a random stimulus (word, image, object, Wikipedia article, news headline) 2. List 10 attributes or associations of that stimulus 3. Force each attribute to connect to the problem 4. At least 2-3 will spark unexpected ideas **Random domain generators:** - Pick a random profession and ask "How would a [chef/architect/jazz musician/surgeon] solve this?" - Pick a random era and ask "How would this be solved in [ancient Rome/the Renaissance/2050]?" - Pick a random constraint and ask "What if we had to solve this with [no money/in 24 hours/for children/underwater]?" ### Engine 7: Mind Mapping & Idea Synthesis **Mind Mapping Process:** 1. **Central node**: Place the core problem/idea at the center 2. **Primary branches**: Radiate 5-8 main themes or dimensions 3. **Secondary branches**: Each primary branch spawns 3-5 sub-ideas 4. **Cross-links**: Draw connections between branches that aren't obviously related 5. **Clusters**: Identify emergent clusters of related ideas across branches 6. **Synthesis**: The cross-links and clusters ARE the creative insights **Output formats:** - Mermaid mindmap diagrams - JSON Canvas (.canvas) files for Obsidian - Structured markdown outlines with cross-references - Knowledge graphs with entity-relationship mappings ## Creative Session Protocol When running a creative session, follow this protocol: ### Phase 1: Diverge (Generate) - Use ALL seven engines on the problem - Quantity over quality — aim for 30+ ideas minimum - No judgment, no filtering, no "that won't work" - Build on others' ideas (yes-and) - Seek the weird, the absurd, the impossible ### Phase 2: Connect (Synthesize) - Create a mind map of all generated ideas - Draw cross-links between ideas from different engines - Look for **unexpected combinations** — ideas from Engine 1 + Engine 5 - Identify **emergent themes** that appear across multiple engines - Ask: "What pattern connects these seemingly unrelated ideas?" ### Phase 3: Converge (Select & Refine) - Evaluate ideas on: novelty, feasibility, impact - Use the "Wow/How/Now" matrix: - **Wow**: High novelty, low feasibility (moonshots) - **How**: High novelty, medium feasibility (worth exploring) - **Now**: Medium novelty, high feasibility (quick wins) - Develop the top 3-5 ideas into fuller concepts ### Phase 4: Articulate (Express) - Present ideas as narratives, not bullet points - Use metaphors and analogies to make abstract ideas concrete - Create visual representations (mind maps, diagrams, sketches) - Connect back to the original problem with a clear "so what" ## Anti-Patterns to Avoid - **Premature judgment**: Never evaluate during divergent phase - **Anchoring on first idea**: The first idea is rarely the best - **Domain fixation**: If all ideas come from one domain, you're not being creative enough - **Complexity bias**: Simple combinations are often the most powerful - **Familiarity bias**: Prefer the unexpected over the comfortable - **Convergent thinking disguised as divergent**: "Brainstorming" that's really just listing obvious options ## Integration with Other Skills This skill works best when combined with: - `thinking-partner` — for challenging assumptions during ideation - `lyn-synthesis-analogy` — for deeper analogical reasoning - `lyn-constraint-creativity` — for constraint-based innovation - `lyn-morphological-triz` — for systematic TRIZ innovation - `lyn-hypotheticals-counterfactuals` — for "what if" exploration - `lyn-mapping-visualization` — for visual thinking - `lyn-knowledge-graph` — for building connection maps - `json-canvas-mindmap` — for creating visual mind maps - `davila-mermaid-diagrams` — for diagramming idea relationships