Skip to main content
idapt
HomeCodeAI ModelsPricing
Sign inStart free
  • Home
  • Pricing
  • AI Models
  • Image models
  • Voice models
  • Video models
  • Rankings
  • New models
  • Model status
  • Multi-Model Chat
  • Voice Mode
  • Voice HUD
  • Web Search
  • Image Generation
  • Video Generation
  • Audio Generation
  • Transcription
  • Drive
  • Credentials
  • Sharing
  • Workspaces
  • Tasks
  • Memory
  • Agents
  • Subagents
  • Automations
  • Skills
  • idapt Code
  • Code Execution
  • Computers
  • Computer Use
  • Computer Assist · Soon
  • Containers · Soon
  • Cloud Computers
  • Local AI
  • AI Gateway
  • API & SDK
  • CLI
  • MCP
  • Tunnels
  • All features →
  • LLM cost calculator
  • Token counter
  • Context window checker
  • Can I run it
  • Model picker quiz
  • Savings finder
  • Video cost estimator
  • Text to speech cost
  • Transcription cost
  • API endpoint tester
  • All free tools →
  • Blog
  • Use cases
  • Comparisons
  • Best of
  • Skills
  • Learn
  • Changelog
  • Help center
  • FAQ
  • Privacy
  • Compare all models
  • Support
  • idapt Code
  • Developers
  • Quickstarts
  • API reference
  • API pricing
  • CLI
  • MCP
  • Downloads
  • Desktop
  • Badges and embeds
© idapt[email protected]TermsPrivacy PolicyLegal noticeReport content
X (Twitter)
Help Center
🔌

idapt MCP Server

Found this helpful? Share it:

idapt runs a hosted MCP server, so any MCP client can drive your workspace through a single tool. Point Claude Code, Cursor, or your own agent at it with an API key and it reads and writes your files, agents, tasks, and more 🔌

How it works

Your client sends JSON-RPC 2.0 requests over Streamable HTTP to https://idapt.app/api/mcp. It authenticates with an idapt API key, passed as Authorization: Bearer YOUR_KEY or x-api-key: YOUR_KEY. There is no OAuth handshake to set up.

The server exposes one tool, idapt. You reach every capability by giving it an action and its arguments:

{ "action": "<resource> <verb>", "args": { ... } }

Only action is required. The command string is the same <resource> <verb>grammar the idapt CLI and the in-app agent speak, so MCP behaves exactly like the rest of idapt. See what an MCP client can do for the call shape and the full surface.

Every call is scoped and workspace-qualified

The server fails closed. Each action maps to a scope on your API key, and an action your key does not carry is denied. user:read is always required. Grant the narrowest set that does the job: see MCP permissions.

Calls address a resource inside a workspace, so the first call to make is { "action": "workspace list" }. It needs no qualifier, confirms the connection, and shows which workspaces the key can reach.

Keys and pricing

MCP works on every signed-in account, including the free tier. You can create up to 1000 API keys at Settings →Developer, so there is no paid-plan requirement to connect a client.

Work done through MCP draws on the same daily allowance, plan usage, and Credit balance as using idapt in the browser. See plans, allowances, and credits.

Related articles

⚙️

Connect your tool

Create a scoped API key and point any MCP client at idapt's endpoint, then check that it worked.

🔧

What an MCP client can do

The one idapt tool, how to call it, how to discover every action from inside your client, and what is off limits.

🔒

MCP permissions

API key scopes, what each one grants, and how to pick the narrowest set for your workflow.

Up next

Connect your tool

Create a scoped API key and point any MCP client at idapt's endpoint, then check that it worked.

Was this helpful?