JasonColapietro/suede-creator-skills / skills/suede-copy
suede-copy
VerifiedUse when asked "rewrite this landing page" or to write, sharpen or retune conversion copy, public explainers and agent-facing docs across surfaces. Suede applies voice, frameworks, labeled variants, answer-first SEO/AEO, evidence boundaries and a 70-point ship score. NOT FOR: a standalone anti-slop pass (use suede-deslop), high-stakes research orchestration (use suede-ship-copy), visual design (use suede-design), or scored SEO audits (use suede-seo-audit).
| File | Kind | Size |
|---|---|---|
| CARD.md | markdown | 4.3 KB |
| SKILL.md | skill md | 16.3 KB |
| agents/openai.yaml | config | 422 B |
| references/edit-passes.md | markdown | 3.1 KB |
| references/email-and-social-formats.md | markdown | 4.9 KB |
| references/frameworks-and-personas.md | markdown | 2.8 KB |
| references/headline-and-cta-formulas.md | markdown | 3.5 KB |
| references/usage-contract.md | markdown | 864 B |
| references/writing-for-agents.md | markdown | 6.4 KB |
references/edit-passes.md
View on GitHub# Suede Seven-Pass Edit
Use when the job is improving copy that already exists. One concern per pass
catches what a single mixed read misses. Every pass keeps the core message,
gives each change a reason, and keeps the author's voice. After each pass,
glance back at the earlier ones so a later edit has not undone them.
## The passes
1. **Clear.** Can a first-time reader follow it? Mark tangled structure,
unclear pronouns, jargon, ambiguity, and missing context before you rewrite.
One main idea per section, addressed to the reader.
2. **Voice.** Read it aloud. Mark shifts between formal and casual, "we"
drifting into "the company", stray technical terms, and a joke that appears
once and never again. Smooth the joins.
3. **Payoff.** After every statement, ask what the reader gets. A feature needs
its consequence spelled out ("which means you can..."). A company milestone
must connect to something the reader receives.
4. **Proof.** List each claim that needs support and check that support sits
nearby: named customer results, data, third-party reviews, guarantees,
logos. Flag "trusted by thousands" and "industry-leading" style lines.
Claims the operator supplied about their own business stay as written and go
to `Claims to verify`. Cut only claims an agent invented or that are
provably false.
5. **Specific.** Replace vague verbs (improve, enhance, optimize), generic
lines, and suspiciously round numbers with real figures, timeframes, and
examples from the source. If a line cannot be made specific, it is usually
filler; cut it. Never invent the number.
6. **Feeling.** Does the reader recognize the before state, the frustration
with the old way, the relief of the fix? Add texture with one small story or
concrete detail. No manufactured urgency, no pressure tactics.
7. **Friction.** Next to each CTA, list every reason to hesitate and confirm
the copy answers it: trial terms, cancellation, guarantee, proof beside the
button, privacy, what happens after the click. Replace "Contact us" with
the real next step.
After pass 7, run passes 6 through 1 once more, quickly.
## Review bench (launch, pricing, and high-traffic pages)
Use after the passes for launch copy, pricing pages, high-traffic landing
pages, and sales pages. Optional for email and ads. Skip for small edits.
1. Seat three to five role reviewers that fit the surface. Landing page:
conversion writer, UX writer, target buyer, brand lead. Email sequence:
lifecycle marketer, copywriter, deliverability reviewer, target buyer.
Long sales page: direct-response writer, skeptical buyer, editor, search
reviewer.
2. Each role scores 1 to 10 in its area and names one specific fix.
3. Fix the lowest scores first, then score again.
4. Stop when every role is at 7 or higher and the average is at least 8.
| Score | Meaning |
|---|---|
| 9-10 | Ready to publish |
| 7-8 | Strong; small fixes |
| 5-6 | Clear gaps; another pass |
| 3-4 | Major revision |
| 1-2 | Rethink the approach |
The bench is a diagnostic. It does not replace the 70-point Ship Gate in
`SKILL.md`, and the roles are simulated lenses, not real people.
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: 472k | 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: 472k | 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: 472k | 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 |