Menu
UX quality bar — Krug's three laws of usability, the eight DevEx first principles, time-to-hello-world tiers, the Goodwill Reservoir, the Pit of Success.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Working with AI coding agents — User Sovereignty, agreement-as-signal, Karpathy/Willison framing, per-model overlays, and the rule to measure your prompt nudges.
Estimate work and scope decisions in the AI-coding era. Boil the Lake, AI compression ratios, three-layer search before building.
Compound knowledge across sessions. Plan-tune dual-track psychographic profile (declared vs behavior), retro philosophy that demotes LOC, the keep-or-toss test for learnings.
Code review, debugging, and incident discipline. Confidence calibration, root-cause iron law, alert-on-changes rule, the pre-existing blame protocol.
When to ship vs not ship. The single ship gate, six auto-decision principles for plan reviews, and the Confusion Protocol for when to stop and ask.
Decide what to build using YC's six forcing questions and the four CEO scope modes. Use before any new feature, product bet, or GTM angle.
Based on SOC occupation classification
| name | ux-bar |
| author | 0xabrar |
| description | UX quality bar — Krug's three laws of usability, the eight DevEx first principles, time-to-hello-world tiers, the Goodwill Reservoir, the Pit of Success. |
What "good" feels like for users. Use when reviewing flows, onboarding, error messages, and dev experience.
Source: gstack design-review/SKILL.md, devex-review/SKILL.md, qa/SKILL.md.
Supporting doctrine:
Every user starts with a finite reservoir of goodwill. Every friction point depletes it. When it hits zero, they leave.
Friction points that drain it:
Goodwill restorers:
"Make the right thing easy, the wrong thing hard."
If users keep doing the wrong thing, the design is wrong, not the users.
For tools where your users are themselves builders:
| Time | Tier | What it means |
|---|---|---|
| <2 min | Champion | Best-in-class. Users will recommend. |
| 2-5 min | Competitive | Acceptable. Most successful tools live here. |
| 5-10 min | Needs Work | Friction is hurting conversion. |
| >10 min | Red Flag | 50-70% abandonment. Fix this before anything else. |
If you can't measure TTHW for your product, you're flying blind on conversion.
At plan stage, score predicted N/10 on each UX dimension.
After ship, do a live audit and score actual.
Flag any dimension where live < plan - 2.
This forces honesty about the gap between what you intended and what you shipped.
When designing for review:
"A designer doesn't hedge, they react. The first impression is gut-reaction first."
Don't ask reviewers what they think — watch them load the page and listen to the first thing they say.
For any nav: if a user lands on a sub-page from search, can they figure out where they are and where to go in 2 seconds?
If not, the nav is broken.
"Every error message must be actionable. The agent should be able to read the error and know what to do next without human intervention."
In an AI-first world, error messages aren't just for humans — they're inputs to a fix loop. "Something went wrong" wastes both human and agent time.
"Repro is everything. Every issue needs a screenshot. No exceptions."
"Depth over breadth — 5-10 well-documented issues > 20 vague descriptions."
"Test like a user. Never read source code."
"Never refuse to use the browser." — even backend-only diffs ship through the browser
| Tier | Time | Scope |
|---|---|---|
| Quick | ~30s | Smoke test, top 5 nav |
| Full | 5-15 min | 5-10 well-evidenced issues |
| Exhaustive | Open-ended | Cosmetic + edge cases |
Triage tier dictates fix scope:
