Brand Voice Critic

Customer-facing

Scores generated drafts against your stored brand voice and gates publication.

What it does

Every draft headed for the editor is auditioned against your brand voice profile. The Critic checks for tone drift, banned phrases, AI tells, and structural slop, returning a score plus a list of fixes. Drafts that score below threshold are flagged before they reach your post editor.

Customize this agent

You can layer additional instructions on top of this agent's built-in system prompt. Your text is appended below the base prompt — it stacks, it doesn't replace. Use it to tell the agent what to emphasise or avoid for your site specifically.

Open Agent settings →

Triggers

→A draft is generated from a journal entry or reservoir
→A promotion is reshaped for X / LinkedIn / Threads
→You click "audit voice" on an existing post

Tools it can touch

•LLM scorer (configurable model)
•Mechanical AI-tell and banned-phrase detectors
•Brand voice profile (per-site, stored in DB)

Features it powers

Post Editor Content Studio

Source

src/lib/brand-voice-critic.ts