Content & Marketing

Give AI your brand voice, not just a prompt.

88% of marketers use AI daily, but only 26% extract real value. The missing ingredient is context.

The problem

Your brand voice lives in a Google Doc no one can find. Tone-of-voice rules are in someone's head. Channel formatting specs are scattered across Slack threads and old decks. Every time AI writes for you, it starts from zero.

How it works

Here's how you build your knowledge graph step by step using the ai CLI.

Add your brand voice guide

Capture the foundational document that defines how your brand speaks. This becomes the root of your content knowledge graph.

ai add "Our brand voice is confident but not arrogant, \
  helpful but not patronizing. We use active voice, \
  short sentences, and concrete examples over abstractions. \
  We never use jargon without explaining it first." \
  --title "Brand Voice Guide" -t policy -d content

Node created as a policy in the content domain.

Add tone of voice rules per audience

Different audiences need different tones. Capture the nuances so AI adapts to the context.

ai add "For enterprise buyers: authoritative, data-driven, \
  reference ROI and scale. For developers: direct, technical, \
  show code examples, skip marketing fluff. For end users: \
  warm, simple, focus on outcomes not features." \
  --title "Tone by Audience" -t policy -d content

Add channel format guides

Each channel has different constraints and conventions. Document them so AI output is publish-ready.

ai add "Blog posts: 1200-1800 words, H2 every 300 words, \
  include a TL;DR, end with a CTA. Social/LinkedIn: \
  max 1300 chars, hook in first line, 3-5 hashtags. \
  Email newsletter: 500 words max, one main CTA, \
  personal tone, plain text friendly." \
  --title "Channel Format Guide" -t process -d content

Add editorial standards

Capture the rules your team follows for consistency: capitalization, terminology, things to avoid.

ai add "Always capitalize Product Name. Never say 'utilize' \
  (say 'use'). Avoid exclamation marks in headlines. \
  Oxford comma always. Dates in 'Month D, YYYY' format. \
  Link to sources for all statistics." \
  --title "Editorial Standards" -t policy -d content

Add your content approval workflow

Document the process so AI tools (and new team members) understand how content moves from draft to published.

ai add "Draft in Google Docs → Self-review with checklist → \
  Peer review (48h SLA) → Stakeholder sign-off for \
  launches → Final proofread → Schedule in CMS → \
  Post to social within 24h of publish." \
  --title "Content Approval Workflow" -t process -d content

Connect everything together

Link related nodes so AI can traverse the graph and pull in all relevant context, not just one document.

ai link abc123 def456 --rel "informs"
ai link abc123 ghi789 --rel "informs"
ai link jkl012 ghi789 --rel "governs"

Now "Brand Voice Guide" informs both "Tone by Audience" and "Channel Format Guide", and "Editorial Standards" governs the channel guide.

Ask AI with full context

When you or an AI agent queries your graph, it assembles all connected knowledge automatically.

ai context "Write a LinkedIn post about our new feature"

Returns your brand voice, LinkedIn format specs, tone rules for the audience, and editorial standards — all in one context package.

Your content knowledge graph is live

Every AI tool your team uses now has access to your brand voice, editorial standards, and channel-specific formatting. No more re-explaining. No more off-brand output.

AI output matches your brand voice from the first draft

New team members write on-brand content from day one

Channel-specific formatting is built into every request

Your standards evolve in one place and propagate everywhere

Start building your knowledge graph

Free during beta. No credit card required.