Knowify OAuth Docs

Appearance

MCP Server ​

The Knowify MCP (Model Context Protocol) server lets AI assistants like Claude interact with your Knowify data using natural language. Instead of writing API calls, you describe what you want and the assistant translates that into structured operations.

What is MCP? ​

Model Context Protocol is an open standard for connecting AI assistants to external tools and data sources. The Knowify MCP server exposes your Knowify resources — projects, clients, invoices, time entries, and more — through MCP resources and (in the future) tools that Claude can use on your behalf.

How It Works ​

Available Resources ​

The MCP server currently serves resources for schema discovery and guides. These let Claude understand the Knowify data model before making requests.

MCP tools (query, create, update, delete) are not yet exposed on this endpoint. See MCP Examples for what will be possible once tools are added.

Supported Resource Types ​

The MCP server supports 21 Knowify resource types. Use MCP resources to discover schema details for each type. See the Resource API reference for the full list.

Connection Methods ​

The Knowify MCP server uses Streamable HTTP transport at:

https://assistant.knowify.com/api/v2/mcp

It is compatible with:

  • Claude Code (CLI) — add to .mcp.json or connect via /mcp
  • Claude.ai — add as a custom connector in Settings
  • Claude Desktop — add via MCP server settings

Scopes ​

MCP enforces the same scope system as the REST API:

  • Read operations require {resource}:read
  • Write operations require {resource}:write
  • Meta-scopes: admin grants all, write grants read+write, read grants read-only

See Getting Started for details on scopes.

Rate Limiting ​

MCP requests are rate-limited at 120 requests per minute per client, matching the REST API. See Errors & Rate Limiting for details.

Next Steps ​