How do I install the Brand Voice Enforce skill?

Run npx skillmds@latest add cleanexpo/brand-voice-enforce in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Brand Voice Enforce safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Brand Voice Enforce?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Brand Voice Enforce free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Brand Voice Enforce?

cleanexpo (@cleanexpo) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Brand Voice Enforce by cleanexpo — Agent Skill

Name: Brand Voice Enforce
Author: cleanexpo

Brand Voice Enforce

Mechanical gate that checks senior-skill output against the brand rules locked in `.claude/memory/ceo-foundation.md` and the gate state in `.claude/memory/verification-gates.md`. Reads both at every invocation. Pass/fail decision per draft. Forwards passing drafts to senior-strategist for final review before CEO batched-review queue.

cleanexpo Updated Sep 17, 2026

File contents

brand-voice-enforce

Mechanical gate. Reads the foundation file at every invocation · derives rules · applies them · passes or fails the draft. Does not embed rule libraries.

When invoked

Before any client-facing artefact lands in the CEO batched-review queue.

What it does

Read ceo-foundation.md sections relevant to the draft's brand + surface
Read verification-gates.md for any gate state the draft references
Check the draft against:
- Universal taboos (foundation Phase 1 + global CEO standing orders)
- Brand-specific taboos (foundation Phase 3.X for the brand)
- Voice tag (foundation Q2.5.5)
- Phase 4 voice amendments
- Aid Rule (foundation Q3.1.1) on RestoreAssist drafts
- Verification-gate tagging (every quantitative claim · [placeholder] or [verified-DD/MM/YYYY])
- Cross-client boundary (foundation Phase 3.4) for CCW drafts
- Hook contract (first-line surfaces): the opening line passes the 6 hook-killer scan — no title-as-hook, greeting, or vague promise (rules owned by senior-copywriter references/hook-craft.md; this gate references them, does not embed them)
Return pass/fail with reasons quoted from foundation

Pass criteria

ALL universal + brand-specific + Phase 4 + verification rules clear. Single rule failure = REJECT with quoted offending string + foundation citation + recommended fix.

Hard rules (the only embedded ones)

No caching of foundation rules. Every invocation re-reads.
No gate flips. Drafts may reference gate state · never flip it (foundation-keeper owns flips).
CEO override scoped to specific artefact. [CEO override · rule X · reason Y] in metadata permits drift on that draft only.
Output object format: { gate_decision, reasons[], warnings[], verification_state[], recommended_revision }.

Versioning

v0.2 (2026-04-27): slimmed from v0.1 · embedded rule libraries removed · delegates to foundation file · token-load reduced ~80%.

---name: brand-voice-enforcedescription: Mechanical gate that checks senior-skill output against the brand rules locked in `.claude/memory/ceo-foundation.md` and the gate state in `.claude/memory/verification-gates.md`. Reads both at every invocation. Pass/fail decision per draft. Forwards passing drafts to senior-strategist for final review before CEO batched-review queue.---# brand-voice-enforceMechanical gate. Reads the foundation file at every invocation · derives rules · applies them · passes or fails the draft. Does not embed rule libraries.## When invokedBefore any client-facing artefact lands in the CEO batched-review queue.## What it does1. Read `ceo-foundation.md` sections relevant to the draft's brand + surface2. Read `verification-gates.md` for any gate state the draft references3. Check the draft against: - Universal taboos (foundation Phase 1 + global CEO standing orders) - Brand-specific taboos (foundation Phase 3.X for the brand) - Voice tag (foundation Q2.5.5) - Phase 4 voice amendments - Aid Rule (foundation Q3.1.1) on RestoreAssist drafts - Verification-gate tagging (every quantitative claim · `[placeholder]` or `[verified-DD/MM/YYYY]`) - Cross-client boundary (foundation Phase 3.4) for CCW drafts - Hook contract (first-line surfaces): the opening line passes the 6 hook-killer scan — no title-as-hook, greeting, or vague promise (rules owned by `senior-copywriter` `references/hook-craft.md`; this gate references them, does not embed them)4. Return pass/fail with reasons quoted from foundation## Pass criteriaALL universal + brand-specific + Phase 4 + verification rules clear. Single rule failure = REJECT with quoted offending string + foundation citation + recommended fix.## Hard rules (the only embedded ones)1. **No caching of foundation rules.** Every invocation re-reads.2. **No gate flips.** Drafts may reference gate state · never flip it (foundation-keeper owns flips).3. **CEO override scoped to specific artefact.** `[CEO override · rule X · reason Y]` in metadata permits drift on that draft only.4. **Output object format:** `{ gate_decision, reasons[], warnings[], verification_state[], recommended_revision }`.## Versioningv0.2 (2026-04-27): slimmed from v0.1 · embedded rule libraries removed · delegates to foundation file · token-load reduced ~80%.

Frequently asked questions