creativity-engine — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited creativity-engine (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Generate novel ideas by combining existing knowledge across domains, then validate them through minimum experiments.
Prompt
Use creativity-engine on this challenge. Generate cross-domain combinations, rank them, and design the smallest test for the best idea.Use Case
Expected Result
Output Example
Verification Case
Verified Effect
"If you own one building block, you can build some cool stuff. If you get a second building block, you can build something more interesting. Get more building blocks, and very rapidly the number of things you can combine them into grows combinatorially or exponentially." — Andrew Ng
The AI Building Blocks you should know:
| Block | What It Does | Combine With |
|---|---|---|
| Prompting | Basic LLM interaction | Any other block |
| RAG | Retrieve and inject context | Knowledge bases, docs |
| Evals | Measure quality | Any output task |
| Guardrails | Safety constraints | Production systems |
| Fine-tuning | Custom behavior | Domain-specific tasks |
| Voice | Speech I/O | Accessibility, hands-free |
| Agentic workflows | Multi-step autonomous | Complex tasks |
| Tool use | External API calls | Real-world actions |
| Embeddings | Semantic similarity | Search, clustering |
| Vector DB | Store/retrieve embeddings | Knowledge management |
Combinatorial ideation process:
Scan the wiki for relevant concepts/entities, then combine them systematically:
[Domain A concept] + [Domain B concept] = Novel ideaExample: AI + Property Maintenance → AI vision diagnosis for apartment repairs
Generate 10-20 ideas across these categories:
The more building blocks you know, the more combinations are possible.
For each promising idea, ask:
For the top 3 ideas, design the cheapest test:
| Idea | Hypothesis | Minimum Test | Success Signal | Cost |
|---|---|---|---|---|
| 1 | If X then Y | Interview 5 users | ≥3 say "I'd pay for this" | Free |
| 2 | ... | Landing page + ads | ≥5% signup rate | $50 |
| 3 | ... | Manual prototype | Works for 1 case | Time |
For Obsidian workflows, save the selected experiment to CREATIVITY_DIR/experiments/ with:
After validation, if the idea survives:
Evaluate each idea:
| Criterion | 1 (Low) | 3 (Medium) | 5 (High) |
|---|---|---|---|
| Value | Nice-to-have | Saves time/money | Creates new capability |
| Difficulty | Requires deep expertise | Feasible with existing tools | Weekend prototype |
| Originality | Common | Unusual in this domain | Never seen before |
CREATIVITY_DIR/experiments/ when working in an Obsidian vault~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.