hashgraph-online/awesome-codex-plugins / plugins/alpoxdev/hypercore/skills/hyper-design
hyper-design
Verified[Hyper] Use this skill when the user wants interface design work that must not look AI-generated - designing a new screen, page, or component, restyling an existing UI, building faux product UI for a hero, feature card, or demo, or auditing, cleaning, or verifying generic template patterns in an interface - driven by a written design direction, a numeric craft floor, static detection, and evidence-separated review, with briefs and reports saved under `.hyper/design/[slug]/`. Do not use it for backend or data work with no visual surface, accessibility-only or performance-only QA, brand strategy with no interface, or a reusable skill folder.
references/ui-copy.md
View on GitHub# Interface Copy
Read when a write mode writes or changes headlines, buttons, labels, empty and error states, or section text, and when an audit finding concerns copy. This covers interface copy only; long-form articles and emails are out of scope. Two rules apply to everything below: never invent a fact, name, number, date, quote, or customer that the source or the user did not supply, and never sterilize the copy into a voiceless default.
## Contents
- 1. Tells
- 2. What not to flag
- 3. Draft, audit, final
- 4. Korean copy
## 1. Tells
| Tell | Before | After |
|---|---|---|
| Empty vocabulary: unlock, elevate, empower, seamless, cutting-edge, revolutionary, next-generation, AI-powered, effortless, robust, game-changer | "Unlock seamless collaboration to elevate your team." | "Work with your team in one shared space." |
| Significance inflation: "the future of", "a new era of", "a testament to" | "Ushering in a new era of work." | "Cuts the time your team spends in status meetings." |
| Trust claims with nothing named: "trusted by thousands", "industry-leading", "SOC 2 compliant" without evidence | "Trusted by thousands of teams." | Name real, permitted customers, or cut the claim |
| Generic CTA: Get started, Learn more, Try now, Explore, Discover | "Get started" | "Create a free workspace", "Book a 20-minute walkthrough" |
| Rule of three forced onto every list | "Innovation, inspiration, and insights." | The real number of items the content has |
| "Not just X, it's Y"; tacked-on fragments ("no guessing") | "It's not just a dashboard, it's a command center." | "The dashboard opens on the jobs that failed overnight." |
| Staccato drama and stage directions ("Read that again.") | "No templates. No defaults. No limits." | One plain sentence |
| False ranges: "from first click to final invoice" | "From onboarding to scale." | "Handles quotes, invoices, and payment reminders." |
| Actorless passive and abstractions with human verbs ("the dashboard understands") | "The pricing page was updated." | "We rewrote the pricing page to show the new plans." |
| Title Case On Every Word in list items and cards | "Instant Alerts Delivered To Slack" | "Instant alerts delivered to Slack" |
| All-caps clauses and bold on every key term | "WE NEED TO MOVE NOW." | Emphasis written into the sentence |
| Em dashes as connectors (`copy.em-dash`) | "Send invoices in minutes β and get paid sooner." | "Send invoices in minutes and get paid sooner." |
| Emoji leading headings and bullets | "π Launch faster" | "Launch faster" |
| Chatbot leftovers and filler: "I hope this helps", "In order to", "It is important to note" | | Cut them |
| Empty, loading, and error text that says nothing: "No data", "Something went wrong" | "No data available." | "No invoices yet. Create one to see it here." / "Payment failed: the card was declined. Try another card." |
## 2. What Not to Flag
Polish, formal vocabulary, one transition word, one em dash in a quoted source, one short emphatic sentence, and curly quotes are not tells on their own. Look for clusters. Leave alone: specific, hard-to-invent detail, mixed feelings, dated references, and varied sentence length; they are signs of a person. A user's own writing sample is a direction: match it, and raise any pattern it uses that `rules/delivery-gate.md` forbids instead of silently keeping or cutting it.
## 3. Draft, Audit, Final
1. Draft the copy in the product's own nouns and verbs.
2. Audit it with two questions: "What here reads as generated?" and "Does it state any fact, name, number, date, or citation not in the source?"
3. Revise for both answers, then search the text for `β` once more.
## 4. Korean Copy
The same tells apply in Korean (μν¬ λ¬Έκ΅¬ "νμ μ μΈ", "λκΉ μλ", "μ°¨μΈλ", "AI κΈ°λ°"; κ·Όκ±° μλ "μμ² κ° νμ΄ μ λ’°"; μΌλ° CTA "μμνκΈ°", "λ μμ보기"). Prefer concrete actions ("λ¬΄λ£ μν¬μ€νμ΄μ€ λ§λ€κΈ°"), keep the product's tone (formal ν©λλ€μ²΄ or friendly ν΄μ체) consistent on one screen, and do not use em dashes as connectors.
## Sources
> Checked 2026-11-07.
| Claim | Source |
|---|---|
| The tell list, the before and after examples, "what not to flag", signs of human writing, and the draft, audit, final loop | <https://github.com/miqdadbadjuber/anti-slop> (commit `9504989`, `antislop-copywriting/SKILL.md` and R-02, R-15, R-16, R-36; MIT, Copyright 2026 Miqdad Badjuber), reduced to interface copy and rewritten for this package |
| Error messages say what failed and how to fix it, without blame | <https://www.nngroup.com/articles/error-message-guidelines/> |
### Evidence grade
A practitioner list without measurement; the source itself says single tells are noise and clusters are the signal.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |