Docs
DocsVoice Guardrails

Voice Guardrails

Last updated: August 10, 2026

Voice Guardrails acts as your brand identity and tone-of-voice auditor. It allows you to set strict directives, forbidden words, and tone archetypes so that all AI-generated content sounds exactly like your brand.

Core Features

  • Rule-Matching Engine: Define exact instructions the AI must follow (e.g., “Use descriptive sensory language”, or “Avoid aggressive sales pitches”).
  • Forbidden Words Scanner: Prevent the AI from using specific off-brand words or phrases.
  • Tone Archetypes: Set percentage blends for various tone archetypes (e.g., 90% Elegant, 10% Playful) to achieve a highly precise writing vibe.
  • Quality Gatekeeper: Acts as a crucial checkpoint in automated production pipelines, ensuring drafts comply with brand guidelines before finalization.
  • AI Audit: The AI can automatically audit your provided text against active guardrails, providing a compliance score and revision feedback.

Workflow

  1. Click New Guardrail to define a brand voice profile.
  2. Set up your Do’s, Don’ts, and Forbidden words.
  3. Set the status to Active to force Eviory AI to pull these rules when drafting.
  4. Use it to audit texts during automated pipeline executions before they hit the Live Work Feed.