Quickstart
The Markup AI API lets you run content agents — Brand Voice, Sentence Clarity, Terminology, and more — against your text and get back flagged issues with suggested fixes. Follow this guide to run your first agent over HTTP.
Using the Markup AI API
Get Your API Key
Create an account, configure your brand voice and style settings in the console, then copy your API key. You’ll send it as a Bearer token: Authorization: Bearer YOUR_API_KEY.
Run Your First Agent
Agents are run by ID, but you never hardcode one — resolve the ID from the agent’s stable name with GET /agents, then POST /agents/{agent_id}/run with your content as text. This example runs Brand Voice (brand_voice), which flags passages that have drifted from your approved brand voice and shares suggestions.
Agents run against the brand voice, personas, terminology, and style guide configured for your organization in the console — there are no per-call style guide IDs. Pass wait=true to block until the run finishes; omit it to get a workflow_id back immediately and track the run with GET /agents/workflows/{workflow_id}.
Personas and Brand Voice Profiles
The persona and brand voice agents tailor their output to a saved configuration that you reference by ID:
- Personas shape feedback for a specific audience or reviewer. Pass a persona’s ID as
persona_idwhen you run the persona agent. - Brand voice profiles capture how your brand should sound. Pass a profile’s ID as
voice_profile_idwhen you run the brand voice agent.
You create and edit both in the Console. To find the IDs to send to the API, list them:
Each entry in the response includes an id — use it as the persona_id or voice_profile_id. The full response shapes are documented under Personas and Brand Voice Profiles in the API Reference.
Want the full agent list and per-agent examples? See Running Agents. For every endpoint, see the API Reference.