From ux-superpowers
Trigger whenever the user says "compile the discovery doc", "synthesize everything into one document", "write the UX discovery document", "produce the final discovery artifact", "consolidate personas jobs journeys and metrics", "extract design principles from our discovery", "surface assumptions and risks", "create the requirements anchor", "wrap up discovery", or any phrase signaling they are ready to assemble prior discovery outputs into a single anchoring document. Always invoke when the user has existing persona, JTBD, journey, telemetry, or research-intake artifacts and wants them fused into one cohesive UX Discovery Document saved to docs/ux-discovery/ with design principles, validated and unvalidated assumptions, risks, technical and business constraints, scope boundaries, and next steps. This is the LAST step in the ux-superpowers discovery pipeline and the canonical handoff point to superpowers:brainstorming — fire it when the user asks "we're done with discovery, what's next", "ready to move to technical brainstorming", or otherwise signals the discovery phase is closing. Do NOT fire for building any single artifact (the dedicated sub-skill owns that), while ux-discover is still orchestrating earlier phases, when the user has no prior discovery artifacts to synthesize, or for general document writing, engineering RFCs, or status reports unrelated to UX discovery.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ux-superpowers:insight-synthesizerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Distill everything into a single document that keeps the team honest.
Distill everything into a single document that keeps the team honest.
Collect outputs from all prior skills:
From all the data, derive 3-5 design principles that should guide every technical decision:
## Design Principles
1. **[Principle Name]**: [One sentence explaining the principle]
- *Because*: [evidence from discovery that supports this]
- *This means*: [concrete implication for what we build]
- *This means NOT*: [what this principle rules out]
Examples:
## Assumptions & Risks
### Validated Assumptions (evidence-based)
- [assumption] — Supported by: [evidence]
### Unvalidated Assumptions (hypotheses)
- [assumption] — How to validate: [suggested method]
- Risk if wrong: [impact]
### Known Risks
| Risk | Likelihood | Impact | Mitigation |
|------|-----------|--------|------------|
| [risk] | High/Med/Low | High/Med/Low | [what to do] |
## Constraints
### Technical Constraints
- [platform, language, infrastructure limitations]
### User Constraints
- [accessibility needs, device limitations, skill levels]
### Business Constraints
- [timeline, budget, compliance, regulatory]
### Scope Boundaries
- **In scope**: [what we're building]
- **Out of scope**: [what we're explicitly NOT building — and why]
- **Future consideration**: [things we might build later]
Assemble the final document with this structure:
# UX Discovery Document: [Project Name]
**Date**: [YYYY-MM-DD]
**Author**: [user + UX Superpowers]
**Status**: [Draft | Reviewed | Approved]
## 1. Problem Brief
[from ux-discover phase]
## 2. Personas
[from persona-builder — include all personas with priority levels]
## 3. Jobs to Be Done
[from jobs-to-be-done — primary and secondary jobs]
## 4. User Journey Maps
[from user-journey — current state and desired state]
## 5. Success Metrics & Telemetry
[from telemetry-designer — north star, health metrics, telemetry plan]
## 6. Design Principles
[from this synthesis]
## 7. Assumptions & Risks
[from this synthesis]
## 8. Constraints & Boundaries
[from this synthesis]
## 9. Research Summary
[from research-intake, if raw data was provided]
## 10. Recommended Next Steps
- [ ] [specific next action]
- [ ] [specific next action]
- [ ] [specific next action]
Save to: docs/ux-discovery/YYYY-MM-DD-<project-name>-discovery.md
Present the document in digestible sections (not all at once). Ask the user to review and approve each major section.
After approval, suggest next steps:
npx claudepluginhub adnanmir123/ux-superpowers --plugin ux-superpowersProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
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.