Access, auth, and limits
Ontraport's MCP server lives at mcp.ontraport.com and is described as a streamable HTTP MCP server. The support docs do not spell out the exact auth handshake, so confirm the token setup in your Ontraport account before wiring it up.
It exposes four capability groups: CRUD across contacts, companies, deals, and custom objects; query tools for discovery, engagement, and account-wide reporting; management tools for tags, automation enrollment, relationships, and tasks; and commerce tools for invoicing, payments, subscriptions, and revenue. The key constraint is that an agent works with a bounded set of records at a time, so plan for segments, not whole-database operations.
Where it works and what to build
Supported clients are Claude Desktop, ChatGPT, and Cursor, plus any MCP-compatible client that speaks streamable HTTP. No CLI is offered, so the MCP connection is the agent path.
Real workflows worth running: ask Claude to tag every contact who bought in the last 30 days and enroll them in a win-back automation; pull a revenue and subscription report across deals without opening the dashboard; or spin up an invoice and payment plan for a new client. Because tools cover commerce as well as CRM, an agent can take an action end to end rather than just surface data.
Which AI clients can use Ontraport MCP?
Not every AI assistant supports MCP natively. Per-client picture for Ontraport:
| Client | Support | Notes |
|---|---|---|
| Claude (Desktop/Web) | Native MCP | — |
| ChatGPT | Native MCP | — |
| Cursor | Native MCP | — |
Does Ontraport MCP cost extra?
The Ontraport MCP server is community-built — free to run, but you still pay for your Ontraport subscription.
- Community MCPs are normally self-hosted from a GitHub repo. There's no extra license fee.
- Your Ontraport account still needs whatever plan is required to access the data the MCP server queries.
- Trade-off: zero MCP cost, but no guarantee it'll keep working when Ontraport changes their UI or API.
Ontraport also publishes a REST API
Ontraport publishes an API, but per-tier gating wasn't extracted automatically.
What you can do with Ontraport from an AI agent
Three shapes of agent workflow. Pick the row that matches how you actually use Claude / Cursor / Codex day to day.
Skip the dashboard, ask Claude
- "What deals stalled this week?" answered in seconds
- Quick one-off questions without leaving the editor
- Pipe Ontraport data into your personal notes / docs without writing a script
Run client analysis at agent speed
- Hand the MCP server to your team — every consultant can ask Claude about any client's Ontraport account
- Multi-account workflows without context-switching the Ontraport UI
- Reports you'd normally spend hours on, done before standup
Production agent workflows
- Embed in your internal Codex / Cursor agents that triage Ontraport data daily
- Combine with other MCP servers in a multi-step agent chain
- Constrain agents to read-only roles to keep production Ontraport safe
Limits and gotchas
- Fallback: the REST API works for any flow the MCP server doesn't cover.
- Rate limits: MCP calls count against your underlying Ontraport API quota. Burst usage from a curious agent can drain a daily allowance fast.
Agent-readiness verdict
Score: 8/10 on MCP-readiness. Ontraport runs a production MCP server (mcp.ontraport.com) over streamable HTTP that works with Claude Desktop, ChatGPT, and Cursor, and the tool coverage spans CRM, automation, and commerce. It loses a point for the bounded-record constraint and the lack of a published auth/transport spec, but for an all-in-one platform this is a real agent surface.
Scored by Joonas (TMB).
Ontraport MCP & API FAQ
Is the Ontraport MCP server official?+
Can I use the Ontraport MCP with Cursor or Codex?+
Does Ontraport support webhooks?+
Sources
- Ontraport official site: https://ontraport.com
Data verified by Joonas on the dates shown. MCP server status auto-rechecked weekly.
10+ years in digital marketing. I review marketing software for AI-stack fit: real pricing, MCP and API support, and how cleanly each tool drops into an AI agent workflow, cross-checked against verified data and real user feedback.