Brand Voice & Content

Codify your voice into rules a machine can check. Voice Extractor + 8-section Voice Spec + On-Brand Writer + 100-point QA Checker + Voice Transfer. Worked example: 22 -> 92.

Lock your brand voice once, from your real writing. Four Claude Skills draft multi-channel content from a single anchor, and a deterministic drift guard scores every draft ON-VOICE / DRIFTING / OFF-BRAND by its weakest dimension — with an anti-slop gate that catches machine-wash the dimensions miss. Grades a draft, not a person; not an AI detector.

Send email that sounds like you, lands in the inbox, and earns the click — honestly. The VOICE standard, voice-locked templates and welcome/re-engagement sequences, humanize prompts, a deliverability QA workbook (SPF/DKIM/DMARC), and a runnable send-readiness linter. No inbox-placement guarantee.

Use AI for speed without sounding like everyone else's AI. A heuristic slop scanner, the SPARK scorecard (Publish / Revise / Rework), a field guide of AI tells with fixes, brand-voice prompts, and a playbook. AI in the decisions, your voice in the words — a quality guardrail, not an AI detector.

Make AI visuals that look like you, not everyone's defaults. Twelve ownable postdigital style recipes (halftone, riso, glitch, dithering, CRT…), a brand-system worksheet that scores each asset On-system / Tweak / Off-brand, and a runnable prompt builder. Real photos for real people; disclose AI; no IP imitation.

Originate short-form video that gets watched: content pillars, a hook engine with a Ship/Tighten/Rework scorer, tight script templates, Kling B-roll prompts, and a cadence you can keep. The creation layer, not just repurposing.

One spreadsheet — pillars, calendar, a repurposing matrix that turns one piece into a week of platform-native posts, and 13 AI prompt blocks with an anti-slop Ship Check.