Skip to content
securemyagent

Articles

Fix: shell, network and file-write with no approval

Cut tool permissions to what the agent needs and require approval for risky calls.

Plain Markdown for agents: /articles/fix-tool-scope.md

Check: tool and permission scope. Controls: ASI02, MCP02.

What is flagged

  • Shell, network and file-write all allowed with no approval.
  • Wildcard OAuth scopes.
  • MCP tools with destructive verbs: delete, drop, send, transfer, exec.

Claude Code

Edit ~/.claude/settings.json. Deny the risky patterns and ask for the rest:

{
  "permissions": {
    "deny": [
      "Bash(rm -rf:*)",
      "Bash(curl:*)",
      "Bash(wget:*)",
      "Read(./.env)",
      "Read(~/.ssh/**)"
    ],
    "ask": ["Bash(git push:*)", "Bash(npm publish:*)"]
  }
}

Codex

codex --ask-for-approval on-request --sandbox workspace-write

Cursor

Turn off auto-run for terminal commands in Cursor Settings. Keep the command allowlist short.

OAuth and API scopes

List the scopes each token has. Replace * and admin with the exact scopes used. Create one token per task.

MCP servers

Read each server's tool list. Remove or disable tools you do not use. Disabled tools cannot be abused.

Verify

npx -y securemyagent checkup

Pro: Guard asks the owner before destructive shell commands, money movement, outbound messages and writes outside the workspace.