Use when you have an idea that wants to become a product too quickly, when you need to break it apart before building, when something feels compelling but fragile, or when you want to find the fracture points before committing. Puts ideas under pressure to see what survives.
How this skill is triggered — by the user, by Claude, or both
Slash command
/judgment-engine-virelan:idea-crucibleThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A skill for breaking ideas apart before building them. Tests what deserves to survive.
A skill for breaking ideas apart before building them. Tests what deserves to survive.
Idea Crucible puts ideas under pressure to see what survives. It:
State your idea in one sentence. What's the essential claim?
What must be true for this to work? Surface 3-5 foundational assumptions.
For each assumption:
What questions are we not asking? What failure modes are we ignoring?
Hold contradictions without resolving them. Some tensions are the answer.
Require at least:
Only assumptions that held under pressure become proto-requirements.
### Core Claim
[One sentence statement]
### Foundational Assumptions
1. [Assumption]
- Fracture points: [Where it breaks]
- Evidence: [What supports it]
- Skeptic says: [Challenge]
2. [Next assumption with same structure]
### Avoided Questions / Failure Modes
- [What we're not asking]
- [What we're pretending won't happen]
### Living Tensions
- [Contradiction we must hold]
- [Trade-off with no clean answer]
### Genuine Unknowns
1. [Something we truly don't know]
2. [Another unknown]
3. [Third unknown]
### What Survived Pressure
[Proto-requirements from assumptions that held]
### Next Pressure Point
[Where to test next]
"I want to run this idea through the Crucible: [idea description]"
"Use Idea Crucible to test this product concept before I write a spec."
"Crucible mode - I think this feels too polished and I don't trust it yet."
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub rjroy/wyrd-gateway --plugin judgment-engine-virelan