HOME SKILLS BLOG GITHUB
// SKILL

BLOG BRAND
ONE EDITORIAL SOURCE OF TRUTH

Stop re-explaining your brand to every command. /blog brand generates two project-root files, BRAND.md and VOICE.md, that every Claude Blog sub-skill auto-loads before it writes or reviews. Audience, positioning, editorial do/don't rules, taboo phrases, competitor differentiation, and voice, captured once and carried everywhere. Free with Claude Blog.

$
/blog brand init

REQUIRES CLAUDE BLOG INSTALLED IN CLAUDE CODE

// WHY IT EXISTS

CONTEXT THAT SURVIVES SESSIONS.

Persona JSON is loaded by some skills and not others. Topic-cluster context lives inside cluster vaults. Competitor positioning lives nowhere. Without a canonical source, each blog command re-derives "what is the brand" from whatever scraps of context it happens to have.

BRAND.md and VOICE.md fix that. They are the editorial equivalent of a persistent PRODUCT.md or DESIGN.md: one canonical source, written to the project root and loaded by the blog orchestrator at the start of every command. Run it once and every sub-skill, from write to audit, works from the same brief.

1
BRAND.md
Who the audience is, what the brand stands for, the editorial do/don't rules, taboo phrases to never use, and how you differ from competitors. The strategic brief.
2
VOICE.md
How the brand sounds, structurally and lexically. Your existing persona JSON re-expressed as readable prose a writer or reviewer can apply directly.
∞
AUTO-LOADED
When present, both files are loaded by the orchestrator before every command. No flag, no re-pasting. Write, rewrite, analyze, and audit all share the same context.
✓
BACKWARD COMPATIBLE
No BRAND.md or VOICE.md? Behavior is unchanged from earlier versions. The files are purely additive. Nothing breaks if you never run the command.
// USAGE

HOW TO RUN IT

Commands

CommandWhat it does
/blog brand initRuns an interactive interview, then writes BRAND.md and VOICE.md to the project root
/blog brand showDisplays the current contents of both files, or reports them missing
/blog brand updateRe-runs the interview with current values pre-filled as defaults

The Interview

The init flow is a short, step-by-step interview. It asks one section at a time and waits for your answer before moving on, covering audience (role, expertise level, the problems readers are solving, common misconceptions), positioning and differentiation, editorial do/don't rules and taboo phrases, and brand voice. If a blog-persona JSON already exists, voice answers are pre-filled from it so you do not repeat yourself.

How Sub-Skills Use It

Once written, the files propagate automatically:

Pairs With blog-persona

blog-persona manages the structured persona JSON; blog-brand re-expresses that persona as readable VOICE.md prose and adds the strategic BRAND.md layer on top. Use persona for machine-readable voice profiles and brand for the human-readable canonical brief.

// FAQ

QUESTIONS ABOUT BLOG BRAND

It generates two project-root files: BRAND.md (who the audience is, what the brand stands for, editorial do/don't rules, taboo phrases, competitor differentiation) and VOICE.md (how the brand sounds, expressed as readable prose). When these files exist, every Claude Blog sub-skill auto-loads them before writing or reviewing, so each command works from one canonical source of brand context instead of re-deriving it.
Three: /blog brand init runs an interactive interview and writes BRAND.md and VOICE.md to the project root; /blog brand show displays the current contents (or reports them missing); /blog brand update re-runs the interview with current values as defaults.
They pair. blog-persona manages the structured persona JSON; blog-brand re-expresses that persona as readable VOICE.md prose and adds BRAND.md for audience, positioning, and editorial rules. If a persona JSON already exists, the brand interview pre-fills voice answers from it.
Yes. When neither BRAND.md nor VOICE.md exists, behavior is unchanged from earlier versions. The files are purely additive context; nothing breaks if you never run /blog brand.
Both are written to the project root so the blog orchestrator loads them at the start of every command. They are the editorial equivalent of a persistent PRODUCT.md or DESIGN.md: context that survives across sessions and propagates to every sub-skill.
// RELATED SKILLS

EXPLORE MORE

WRITING PERSONAS

Manage structured persona JSON and voice profiles per brand. blog-brand re-expresses persona as VOICE.md prose.

PERSONA MANAGER →

BLOG STRATEGY

Niche positioning, topic ideation, and content pillar architecture, anchored to the same BRAND.md audience and positioning.

STRATEGY PLANNER →

BLOG WRITER

Drafts full articles that apply the voice and respect the taboo phrases captured in your brand context automatically.

CONTENT WRITER →

VIEW ALL 30 SKILLS →

CAPTURE YOUR BRAND ONCE.
USE IT EVERYWHERE.

$
git clone --depth 1 https://github.com/AgriciDaniel/claude-blog.git && bash claude-blog/install.sh
VIEW ON GITHUB ALL SKILLS >