macOS Gatekeeper warning for bundled rg in Codex extension on Antigravity

What version of the IDE extension are you using?

openai.chatgpt-26.506.31421-darwin-arm64

What subscription do you have?

plus

Which IDE are you using?

Antigravity

What platform is your computer?

Darwin 24.6.0 arm64 arm

What issue are you seeing?

When using the ChatGPT extension in Google Antigravity on macOS (Apple Silicon), macOS Gatekeeper displays the following warning during some Codex tasks:

“Apple cannot check ‘rg’ for malicious software.”

Clicking Done dismisses the warning, and Codex continues the task and completes it successfully.

Environment

• macOS (Apple Silicon)
• Google Antigravity (latest release)
• ChatGPT extension: openai.chatgpt-26.506.31421-darwin-arm64

Additional information

• The bundled rg binary runs normally when executed directly:

~/.antigravity/extensions/openai.chatgpt-26.506.31421-darwin-arm64/bin/macos-aarch64/rg --version

• Homebrew rg also runs normally.
• syspolicyd logs contain:
• GatekeeperPolicyScanError Code=-67018
• Code did not match any currently allowed policy
• team: (null), id: rg

What steps can reproduce the bug?

1. Open Google Antigravity.
2. Open a workspace with the ChatGPT extension enabled.
3. Ask Codex to create or write a Notion document.
4. During the task, macOS displays the Gatekeeper warning:

“Apple cannot check ‘rg’ for malicious software.”

1. Click Done.
2. Codex continues and completes the task successfully.

What is the expected behavior?

_No response_

Additional information

_No response_