PlatformMCPBeta
Connect Claude, ChatGPT or any MCP assistant to Centium, then ask about your brands in plain language.
https://api.centium.ai/mcpWorks with any MCP app, including
Turn the searches AI models run into a blog post.
Using Centium, look at our query fan-out. Pick the searches that come up most, then write a blog post that answers them in plain language. List which searches each section covers.
Reads: Query Fan-Out, Fan-Out Themes
Find the most-cited writer who hasn’t covered you yet, and draft the pitch.
Using Centium, find the writer AI cites most in our categories whose cited articles don’t mention us yet. Summarize what they cover, then draft a short pitch email with an angle that fits their beat.
Reads: Influential Voices, Citations
Turn your trend data into slides for the next meeting.
Using Centium, create a 5-slide presentation on our AI visibility over the last six months: the overall trend, each AI model, our top categories, how we moved against competitors, and the next steps from our game plan.
Reads: Recommendation Trends, Model Visibility, Competitor Comparison, Visibility Optimization
https://api.centium.ai/mcp
Sign in with your Centium account, then ask your agent to act on your Centium data.
Centium measures how AI models talk about your brand and what you can do about it. The MCP puts that data inside Claude, ChatGPT and the other assistants your team uses every day.
Starting at $99/month
MCP (Model Context Protocol) is the open standard AI assistants use to connect to outside tools and data. The Centium MCP server lets Claude, ChatGPT and other assistants read your Centium dashboards when you ask them a question.
Chat apps like Claude, ChatGPT, Gemini and Grok, coding tools like Claude Code, Codex and Cursor, and agents like Grok Bot and OpenAI dots, plus any app that supports remote MCP servers with sign-in. They all use the same address, https://api.centium.ai/mcp. See the setup steps for each.
No. You sign in with your Centium account when your AI app asks. There are no keys to copy or manage.
Anyone with access to a Centium brand dashboard, from managers to viewers. Each person signs in with their own Centium account and sees only the brands that account can open, including brands shared through a group. Share links do not grant access.
The same data as your dashboard and its export: visibility, competitors and position, prompts, sources, perception, query fan-out and your game plan. See what it can read.
As fresh as your dashboard. When a new run finishes, your AI app sees it on the next question.
Yes. Every past run stays available by date, so you can ask how anything changed between two runs.
No. The connection is read-only. Your assistant can read dashboard data and past runs. It cannot edit prompts, models, settings or team access.
Sign-in uses OAuth, the connection is read-only, and Centium checks your access on every request. Remove someone from a brand and their AI app loses access right away.
Centium sends your AI app only the data it asks for. How that app stores or uses it follows your agreement and data settings with that provider.
Open Integrations in your dashboard menu and click Disconnect, or remove the connector in your AI app’s settings. Access ends right away.
Start with the prompt library, or ask your AI app to act on your Centium data.
Yes. It is in beta while we learn how teams use it. Tell us what you would like it to do at hello@centium.ai.