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
🔒

MCP permissions

Found this helpful? Share it:

An API key carries scopes that decide which resources a client can reach. Access fails closed, so grant only the scopes a workflow needs 🔒

How scopes work

A scope is a resource name with a read or write level, written drive:read or drive:read,write. The server denies any action a key is not scoped for, and it denies any action with no scope mapping at all. Every key must include user:read.

Scopes

Scope

Read grants

Write grants

user

Your profile (required on every key)

Update account preferences

workspace

List and read workspaces

Update workspace settings

drive

Read, search, and list files

Create, edit, move, and delete files

chat

Read chats and messages

Create chats and send messages

agents

List and read agents

Create and update agents

tasks

List and read tasks

Create, update, comment, and delete tasks

computers

List computers and their status

Run commands, transfer files, manage tunnels and users

memory

Recall an agent's memory notes

Remember and forget memory notes

automations

List automations and their runs

Create, fire, and rotate automations

credentials

List workspace credentials (metadata only)

Create, rotate, and delete credentials

notifications

Read your notifications

Send notifications to members

completions

List models

Run model inference (text, image, speech)

Further scopes cover realtime, members, billing, and the repos, functions, and CI surfaces. A key can only exercise a scope its owner could exercise in the app.

Pick the narrowest set

Some starting points:

  • Read files and docs: drive:read, workspace:read, user:read.

  • Full coding workflow: drive:read,write, chat:read,write, tasks:read,write, agents:read, workspace:read, user:read.

  • Computer operations: computers:read,write, drive:read, workspace:read, user:read.

Mint a separate key per client so you can rotate or revoke one without touching the others. See API keys and scopes.

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.

🔌

idapt MCP Server

Connect Claude Code, Cursor, Codex, and any other MCP client to your idapt workspace.

Up next

Connect Claude Code

Drive your idapt workspace from Claude Code over the CLI or MCP, with a shared memory and agent identity.

Was this helpful?