tools/sage-claude-plugin/skills/ux-discovery/SKILL.md
User research and context gathering — understands who users are, what they do, and why
npx skillsauth add xoai/sage ux-discoveryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Gathers user context to ground specifications in real user behavior rather than
developer assumptions. Runs alongside quick-elicit at the elicitation phase.
Add these UX-focused questions to the elicitation, woven into the normal
quick-elicit flow. Don't extend the time — make the existing questions sharper.
Ask about the user's context (pick 2-3 most relevant):
Capture in the user-context-notes artifact:
Run the complete UX discovery process. This is a structured investigation that produces artifacts feeding directly into specification.
Review what's known about the users from:
Apply The Mom Test principles to structure the inquiry:
Produce: research findings summary
From the research, develop 2-3 behavioral personas:
[Name] — [Role]
"[Quote capturing their attitude]"
GOAL: What they're trying to accomplish
CONTEXT: When, where, on what device
PAIN: What frustrates them about current solutions
BEHAVIOR: How they approach problems
FREQUENCY: How often they do this
Designate one as the PRIMARY persona (design for this person first). Optionally designate a NEGATIVE persona (explicitly NOT designing for).
Produce: persona-profiles artifact
Map the primary persona's full journey:
Produce: journey-map artifact
From the journey map, extract the top pain points ranked by severity:
These feed directly into the specification as requirements and acceptance criteria.
Load from references/ as needed:
user-behavior-model.md — How users actually behave (Krug, Norman)user-research-conversations.md — Mom Test rules for useful conversationspersona-development.md — Persona construction frameworkjourney-mapping.md — Journey mapping components and processdevelopment
Branch-per-initiative git discipline for all delivery workflows. Defines branch naming by workflow, the propose-confirm creation protocol, dirty-tree and detached-HEAD handling, the always user-gated merge protocol, worktree support for parallel sessions, and abandonment cleanup. Activates only in git repositories — silently inactive everywhere else. Use when starting /build, /fix, /architect, or /build-x at Standard+ scope, when resuming an initiative, when offering a merge at a completion checkpoint, or when the user wants a second concurrent initiative.
development
Drives task-by-task execution from an approved plan with quality gates between each task. Reads the plan, finds the next incomplete task, dispatches implementation, validates, updates progress, and continues. Use after a plan is approved and the user says "go", "start building", "execute the plan", or "implement the feature".
testing
Preserves and restores context across agent sessions using plan file checkboxes as source of truth. Use when starting a new session, resuming previous work, ending a session, or when the user says "continue from last time", "what was I doing", or "save progress".
tools
Captures agent mistakes, corrections, and discovered gotchas so they are not repeated. Use when: (1) a command or operation fails unexpectedly, (2) the user corrects the agent, (3) the agent discovers non-obvious behavior through debugging, (4) an API or tool behaves differently than expected, (5) a better approach is found for a recurring task. Also searches past learnings before starting tasks to avoid known pitfalls. Activate alongside the sage-memory skill — they share the same MCP backend but serve different purposes (sage-memory = codebase knowledge, sage-self-learning = agent mistakes and gotchas).