What is MCP (Model Context Protocol)
Model Context Protocol (MCP) is an open standard that lets AI systems securely connect to external APIs, data sources, and tools.
It defines how models discover, describe, and call external actions in a consistent format.
Instead of building one-off integrations, MCP lets everything plug into the same interface.
Each connected system is an MCP server exposing “tools” (actions), “resources” (data), and “prompts” (templates).
These can be called by AI clients or automation frameworks that support MCP.
Communication uses JSON-RPC 2.0 over HTTP, SSE, or stdio.
Authentication and access control are handled by each server.
Major AI platforms like OpenAI, Anthropic, and Microsoft already support MCP.
This standard makes integrations faster, safer, and easier to maintain.
Below is a list of the MCP servers we support.
Available MCP servers
| MCP | Docs URL | MCP URL | Auth |
|---|---|---|---|
| Partner API doc | https://docs.partner-api.monta.com/docs/mcp | https://docs.partner-api.monta.com/mcp | No auth needed |
| Partner API | https://partner-api-mcp.monta.app | https://partner-api-mcp.monta.app/sse & https://partner-api-mcp.monta.app/mcp | Partner API credentials required |
| Knowledgebase | https://knowledgebase-mcp.monta.app | https://knowledgebase-mcp.monta.app/sse & https://knowledgebase-mcp.monta.app/mcp | No auth needed |