Menú
Brand voice checklist for every external-facing content review. Apply during review_content tasks before scoring the submission.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
| name | brand-enforcement |
| description | Brand voice checklist for every external-facing content review. Apply during review_content tasks before scoring the submission. |
| metadata | {"version":"1.0.0","type":"always-active"} |
The canonical brand voice is defined in prompts/brand-voice.md. This skill is the
operational checklist Reviewer runs on every submission.
Every piece of content must be all four:
| Attribute | Sounds Like | Doesn't Sound Like |
|---|---|---|
| Direct | "12 agents. 6 departments. Zero humans in the loop." | "We're excited to announce our innovative AI-powered solution…" |
| Technical but accessible | "Event-driven architecture means agents react to what happens, not what's scheduled." | "Our proprietary synergistic platform leverages cutting-edge paradigms…" |
| Confident | "Open source because agent infrastructure shouldn't be a walled garden." | "The ONLY solution for enterprise AI orchestration!" |
| Community-first | "Fork it. Break it. Ship something better. That's the point." | "Join our exclusive early access program…" |
If the content fails any attribute → reduce score to <90. If it fails two or more →
reviewer:flagged with specific rewrite suggestions.
The following marketing-speak is banned. Flag and rewrite.
| Banned | Why | Use Instead |
|---|---|---|
| "revolutionary" | Empty superlative, hype-marketing | Describe what it does |
| "game-changing" | Same | Describe the actual change |
| "cutting-edge" | Cliché | Name the specific technology |
| "next-gen" | Lazy | Name the generation if relevant |
| "synergy" / "synergistic" | Corporate filler | Remove |
| "leverage" (as verb) | Vacuous | "use" |
| "solution" (generic) | Meaningless | Name what it is (framework, library, agent) |
| "AI-powered" (as differentiator) | Redundant for this product | Name the specific capability |
| "We're excited to…" | Cliché opener | State the thing directly |
Public-facing content MUST use this framing order:
NOT the reverse. Benefit-first copy reads as marketing. Mechanism-first reads as engineering confidence.
Bad (benefit-first):
"Transform your organization with next-gen AI orchestration that unlocks unprecedented productivity."
Good (mechanism-first):
"YClaw runs agents in real departments with event-driven coordination and approval gates. Extracted from a production system that ran 12 autonomous agents for over a year. Open source under AGPL-3.0."
| Form | Where it's correct |
|---|---|
| YClaw | Default for all external / user-facing content — name of the product |
| YCLAW | Repo name + internal documentation + path names only (e.g., YClawAI/YClaw) |
| yclaw / Yclaw / YClaw™ | ❌ Never |
Fail-closed rule: if you see YCLAW in user-facing content, flag it. If you see
yclaw or Yclaw anywhere, flag it.
The package / GitHub repo name is YClaw. The brand in prose is YClaw. The internal
long-form / repo identifier is YCLAW.
Never use future tense for shipping announcements. If it ships, it IS shipped, not "will be shipped":
If a feature is NOT shipped:
Every external-facing post MUST include at least one of:
https://github.com/YClawAI/YClawhttps://yclaw.aiLinks must be:
?utm=* unless Reviewer/Ember explicitly approves)YClawAI/YClaw (not forks, not mirrors)From prompts/brand-voice.md:
Apply this to every review:
| Score | State |
|---|---|
| 90–100 | All four voice attributes, no banned terms, framing correct, links present |
| 70–89 | One minor voice slip OR one stylistic issue (e.g., missing link, mild hype) |
| 50–69 | Two+ voice issues OR one structural issue (framing reversed, YClaw capitalized wrong) |
| 0–49 | Banned term used, future tense on shipped feature, legal-rail violation |
Anything ≤49 → automatic BLOCK with specific rewrite. Legal flag at any score → automatic BLOCK regardless of voice.
claims-risk skill (separate, higher priority).channel-standards skill.oss-legal-guardrails skill.Detect and prevent product-specific terminology contamination across skills/ and prompts/. Run after any bulk content update, skill import from another project, or agent config change. Applied as a shared skill across all agents.
When and how Architect delegates tasks to AO, Mechanic, and Designer.
Checklist and rules for reviewing deployment requests. Architect assesses but NEVER executes.
Rules for triaging new issues, classifying severity, and routing to the correct agent.
How to classify and respond to CI/pipeline failures.
Rules for detecting and managing stale issues and PR reviews.