← ClaudeAtlas

review-voicelisted

Review a content file against content/VOICE.md for what the lint cannot judge - claims, numbers, regulation naming, rhythm, structure - after quoting the lint's own findings on it, and report HIGH / MEDIUM / LOW with a verdict. Use when asked to brand-review, voice-check, lint, audit or read a draft or a page for voice, claims, AI-tic prose or readiness, or "is this on-brand" / "is this ready to publish".
HKamkar/agentic-cms · ★ 2 · Code & Development · score 71
Install: claude install-skill HKamkar/agentic-cms
# Review voice The repo is the current directory. The file is the one the user named (`content/blog/<slug>.md`, `content/pages/<slug>.yaml`, a FAQ set); if none, ask. One file per run: a report on every draft at once is too long to act on. ## Read first `content/VOICE.md` in full: the reader, the non-negotiables, the claims, the writing rules, and the fenced block (the exact rules the lint runs). `content/blog/_template.md` for a post's structural fields. Then the file. ## Passes 1. **The lint's lines.** Run `pnpm content:lint` and quote every line for this file verbatim: those are the exact rules (banned words and patterns, the brand's spelling and mark, model and cloud names, em dashes, alts, counts, headings, images, dates) and are not re-judged; a `FAIL` is a HIGH finding as it stands. If the command cannot run (no permission in this session), say so on the `Lint:` line and check the block's exact rules by hand, marked as a hand-check. 2. **Claims** (HIGH): compliance or certification language `VOICE.md` forbids, however phrased; numbers without a documented source; a customer, logo, quote or case study without approval; a roadmap promise; anything `VOICE.md` says is never said (offers named wrongly, the incumbent framed wrongly, entities it never names). 3. **Naming** (MEDIUM): a regulation or standard abbreviated before its full name; the offers and the category spelled as `VOICE.md` spells them; the calls to action as it prescri