Otto MCP
Otto's Model Context Protocol (MCP) server enables businesses to automate processes with intelligent AI-powered digital workers. Connect AI agents and systems to streamline workflows, boost productivity, and transform operations.
Install Otto MCP Server in Cursor
Prerequisites
- Cursor IDE installed (latest version)
- Otto API Token with appropriate scopes
- For local installation: Docker installed and running
Remote Server Setup (Recommended)
Uses Otto's hosted server at https://api.otto.dev/mcp/. Requires Cursor v0.48.0+ for Streamable HTTP support. While Cursor supports OAuth for some MCP servers, the Otto server currently requires an API Token.
Install steps
- Click the install button above and follow the flow, or go directly to your global MCP configuration file at
~/.cursor/mcp.jsonand enter the code block below - In Tools & Integrations > MCP tools, click the pencil icon next to "otto"
- Replace
YOUR_OTTO_TOKENwith your actual Otto API Token - Save the file
- Restart Cursor
Streamable HTTP Configuration
{
"mcpServers": {
"otto": {
"url": "https://api.otto.dev/mcp/",
"headers": {
"Authorization": "Bearer YOUR_OTTO_TOKEN"
}
}
}
}Actions
Available MCP actions and their usage
Tools
•
list_automations- List currently active automations•
run_automation- Trigger the execution of a specific automation○
automation_id: The unique identifier of the automation to run (string, required)○
inputs: Optional input parameters for the automation (object, optional)•
query_audit_trail- Query audit trail data, including errors, exceptions, and successful runs○
automation_id: The automation ID to query audit trail for (string, required)○
status: Filter by execution status (success, error, running) (array, optional)○
limit: Maximum number of results to return (number, optional)○
start_date: Start date for the query range (ISO 8601 format) (string, optional)○
end_date: End date for the query range (ISO 8601 format) (string, optional)Roadmap
Upcoming Features
Action support for creating or pausing automations
Webhook and event-streaming for real-time automation status
LangChain, OpenAI Assistants, and Zapier integrations