Permissions & settings
What to let an agent do, approval gates and per-agent settings walkthroughs.

Least Privilege for Personal AI Agents: Connect Less, Grant Narrower
Least privilege for AI agent permissions means connecting only the apps a task needs, choosing read over send, and using short approvals instead of permanent ones.
Read Least Privilege for Personal AI Agents: Connect Less, Grant Narrower
When Your Agent Hands Back Control: Takeover, Passwords, 2FA and CAPTCHAs
If an AI agent asks for your password in chat, stop. Legitimate agents hand back control or use a secure form for passwords, 2FA codes and CAPTCHAs, never a chat message.
Read When Your Agent Hands Back Control: Takeover, Passwords, 2FA and CAPTCHAs
Why "Always Allow" Is the Riskiest Button in Your AI Agent
An Always Allow AI agent permission turns one approval into standing access, so a later mistake or hidden instruction runs unchecked. Use one-time approvals instead.
Read Why "Always Allow" Is the Riskiest Button in Your AI Agent
Claude Duet Approval Settings: What Always Asks, What Never Runs
Claude Duet approval settings: Duet always asks before paying, deleting an order or creating an account, and refuses trading or moving money even if you approve.
Read Claude Duet Approval Settings: What Always Asks, What Never Runs
Grok Bot Computer Access Settings: Ask Every Time, Always Allow and the Shared Computer
Grok Bot computer access settings: keep Settings, Computer on Ask every time, use Ask first rules for sending and buying, and know that all your Bots share one computer.
Read Grok Bot Computer Access Settings: Ask Every Time, Always Allow and the Shared Computer
Meta Muse Safety Settings: Approval Scopes, Connectors and Revoking Access
Meta Muse safety settings come down to three levers: approval scopes in the app, per-connector access levels, and disconnecting apps. Here is how to set each safely.
Read Meta Muse Safety Settings: Approval Scopes, Connectors and Revoking Access
Setting Up dots Custom Rules Safely: The Four Modes and Sensible Defaults
dots Custom Rules settings give each action one of four modes: act, act if pre-approved, ask first, or hand off. Start with ask first for sending and spending.
Read Setting Up dots Custom Rules Safely: The Four Modes and Sensible Defaults
OpenClaw Sandboxing and Command Approvals: Turning On the Safety Rails
OpenClaw sandbox settings are off by default in 2.0. Turn on the Docker or Podman sandbox, limit workspace access, and use allowlist or ask mode for command approvals.
Read OpenClaw Sandboxing and Command Approvals: Turning On the Safety Rails
Read-Only vs Send Access: Choosing Email and Calendar Scopes for an AI Agent
Give an AI agent read-only email access first: it can summarise and draft without sending. Add send access only with approval prompts, and keep calendar edits narrow.
Read Read-Only vs Send Access: Choosing Email and Calendar Scopes for an AI Agent
What to Let a Personal AI Agent Do (and What to Keep for Yourself): A Risk Ladder
What should you let an AI agent do? Start with read-only research and drafts, add sending with approval, and keep money moves, passwords and deletions for yourself.
Read What to Let a Personal AI Agent Do (and What to Keep for Yourself): A Risk Ladder