What is MCP?
MCP, or Model Context Protocol, is a standard way for AI clients to connect to external products. Instead of asking an assistant to guess how Ocoya works, you connect the assistant to Ocoya MCP and it can discover typed tools for supported actions. Ocoya MCP exposes workspace, publishing, brand, Studio, and workflow tools after the user approves OAuth access.AI client connects
Claude, ChatGPT, Codex, VS Code, or another MCP-compatible client connects to
https://mcp.ocoya.com over Streamable HTTP MCP.User approves access
Ocoya opens an OAuth approval flow where the user signs in, reviews permissions, and selects the workspaces the client can access.
Why use MCP for Ocoya?
MCP is useful when you want an AI assistant to perform product actions directly and safely:- The assistant can discover Ocoya tools instead of inventing API calls.
- Users approve access through OAuth instead of copying API keys.
- Workspace access is scoped to the workspaces selected during approval.
- Tool calls return structured data that clients can use in follow-up actions.
What Ocoya exposes
Ocoya MCP gives approved clients tools for:| Area | What the assistant can do |
|---|---|
| Account and workspace | Identify the authorized user and list approved workspaces. |
| Social profiles | List connected channels and start new social connection flows. |
| Posts | Create, inspect, update, schedule, publish, and delete post groups. |
| AI drafts | Generate editable draft posts from prompts with optional brand and visual context. |
| Brand context | List brand kits, hashtag libraries, and Studio templates. |
| Workflows | List, inspect, enable, disable, and run workflows. |
Authentication model
Ocoya MCP uses OAuth. A client cannot use Ocoya MCP with an Ocoya API key. When the client first needs an Ocoya tool, it opens Ocoya in the browser. The user signs in, reviews requested permissions, selects allowed workspaces, and approves the client. The client receives a short-lived access token and can refresh access according to the OAuth flow. If a workspace was not selected during approval, tools cannot access that workspace.MCP vs REST API
Use MCP when an AI client should act on the user’s behalf through natural language and tool calls. Use the REST API when you are building a backend integration, scheduled job, or application that calls Ocoya endpoints directly.Next
Connect Ocoya MCP
Choose your MCP client and add the Ocoya MCP server.
MCP tools
Review all available Ocoya MCP tools and parameters.