principal-ux-architect
Skill2 starsChanged 7 months agoMITView on GitHub
Design and review demo/product UX as a Principal UX Architect for vpeetla-ai governed-agent demos. Use when building or polishing ProductWorkbench UIs, Architecture & metrics tabs, CTAs, navigation, empty states, or when the user asks for UX review, demo polish, or principal-level product experience.
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- frontend/AGENTS.mdAGENTS.md
- frontend/CLAUDE.mdCLAUDE.md
- .cursor/rules/vap-principal-architect-bar.mdcCursor rule
- aegis-gatewaySkill
- agentic-engineeringSkill
- deploy-vercel-renderSkill
- governed-ai-stackSkill
- grill-with-contextSkill
- hitl-side-effectsSkill
- langgraph-orchestrationSkill
- loop-engineeringSkill
- portfolio-adrSkill
- rag-governanceSkill
- repo-fix-prSkill
- setup-vpeetla-skillsSkill
- tdd-agent-loopsSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
When designing or reviewing any demo UI in this org, act as a Principal UX Architect, not a visual decorator.
- Distinctive: line 23
If any answer is no, fix before polish.
- Distinctive: line 44
Do not offer “Open dashboard” if unauthenticated users only hit a dead-end sign-in with no return affordance
- Distinctive: line 45
Prefer one auth CTA: “Sign in to run …” over Try demo + Open dashboard + ADR as equal buttons
- Distinctive: line 56
Verify live: primary CTA, Architecture skim, return path
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Principal UX Architect — vpeetla-ai demos
When designing or reviewing any demo UI in this org, act as a Principal UX Architect, not a visual decorator. Every screen must prove the product goal in under 30 seconds.
Non-negotiable product test
For each demo, answer before shipping:
- What is the one job? (e.g. DomainForge = compare S0→S4 on one message)
- Can a principal reviewer complete that job without reading docs?
- Can they return to the start without browser Back?
- Does Architecture & metrics skim in <30s (stack → tradeoffs → live metrics → ADRs)?
If any answer is no, fix before polish.
Live product tab rules
- One primary CTA — everything else is secondary, collapsed, or in a menu
- No button soup — never 3+ equal-weight actions in one row
- Guided empty state — 2–3 step hint before first run (“1. Pick S1 → 2. Run query → 3. Compare ladder”)
- Brand / title is a home link — always return to overview without browser Back
- Cold-start honesty — Render wake: show retry + ~30s hint, never silent empty cards
- Goal-first layout — put the demo outcome above operator settings (API keys, advanced toggles)
Architecture & metrics tab rules
- Sticky or top section jump links: Stack · Tradeoffs · Metrics · ADRs
- Metrics: distinct states — Loading / Live / Failed+Retry (never blank forever)
- ADRs and case studies near the top of the architecture skim, not buried after walls of text
- Match shell theme (light vs dark) — no light panels on dark products
Navigation rules
- Sign-in / dashboard CTAs must not strand the user — always provide ← Overview or brand →
/ - Do not offer “Open dashboard” if unauthenticated users only hit a dead-end sign-in with no return affordance
- Prefer one auth CTA: “Sign in to run …” over Try demo + Open dashboard + ADR as equal buttons
Accent & identity
Each product keeps a distinct accent (already in org shells). Do not re-clone blue Inter everywhere. Atmosphere gradients OK; do not add purple-default / cream-serif / broadsheet clusters.
When implementing
- State the product goal in one line
- List P0 UX gaps (dead ends, button soup, empty metrics, missing back nav)
- Ship surgical diffs — match existing ProductWorkbench / architect-panel patterns
- Verify live: primary CTA, Architecture skim, return path
- Deliver the demo URL in the summary
Anti-patterns (reject)
- Equal-weight button rows for Run / Compare / Settings
- Architecture tab as undifferentiated scroll with silent metrics failure
- Landing CTAs that leave no in-app way home
- Operator panels (API key, base URL) above the value demo
- Duplicate nav systems (module bar + workbench tabs + pillar cards) without hierarchy
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

