Use when the question is too big to solve but too important to guess, when you need thinking space that sharpens structure, when answers feel fluent but hollow, or when each thought adds complexity without traction. Generates multiple distinct reasoning paths with assumptions and risks visible.
How this skill is triggered — by the user, by Claude, or both
Slash command
/judgment-engine-virelan:exploratory-reasoningThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A skill for breaking open big questions and returning multiple paths with their assumptions and risks visible.
A skill for breaking open big questions and returning multiple paths with their assumptions and risks visible.
Exploratory Reasoning generates multiple distinct reasoning paths for complex questions:
Take your question and reframe it for clarity. Sometimes the reframing itself reveals assumptions.
Create 3-5 genuinely different approaches. Not:
But truly different lenses or frameworks for thinking about the problem.
Assumptions: What must be true for this path to work?
Risks: Where does this approach break down?
Right if: Under what conditions would this be the correct path?
Questions that test across all paths:
### Root Question
[Restated and sharpened version of your question]
### Path 1: [Approach Name]
**Core idea**: [What this path prioritizes]
**Assumptions**:
- [What must be true]
- [Another assumption]
**Risks**:
- [Where this breaks]
- [What this ignores]
**Right if**: [Conditions under which this path works]
### Path 2: [Different Approach]
[Same structure]
### Path 3: [Yet Another Approach]
[Same structure]
### Cross-Path Prompts
- What breaks first in each path under pressure?
- What does each path pretend not to care about?
- [Additional testing questions specific to these paths]
Question: "How should we prioritize technical debt?"
Path 1: User Impact First
Path 2: System Health First
Path 3: Team Capacity First
"Run Exploratory Reasoning on: How should we structure our testing strategy?"
"I need different thinking paths for this architecture decision."
"Exploratory mode - I keep circling the same solution and need genuinely different approaches."
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