Skip to content

fix(codex): restart the daemon without Wizard's TCC attribution - #1

Open
razmser wants to merge 1 commit into
devfrom
razmser/responsibility
Open

razmser wants to merge 1 commit into
devfrom
razmser/responsibility

Conversation

@razmser

@razmser razmser commented Oct 26, 2026

Copy link
Copy Markdown
Collaborator

macOS charged the Codex daemon that Wizard restarted, and everything Codex ran in it, to Wizard, so users saw "CoreInfra Wizard tried to access data from other apps" notifications.

  • When Wizard is its own responsible process (the app), spawn the restart with responsibility disclaimed. A CLI run from a terminal keeps the terminal's attribution.
  • Use daemon bootstrap instead of restart, so a running updater is replaced too. An existing updater may still be attributed to an older Wizard, and it keeps proxy variables that were removed from .env. Preserve remote control, which bootstrap would otherwise reset.
  • Skip the restart unless daemon version reports a managed daemon.

macOS charged the Codex daemon that Wizard restarted, and everything
Codex ran in it, to Wizard, so users saw "CoreInfra Wizard tried to
access data from other apps" notifications.

- When Wizard is its own responsible process (the app), spawn the
  restart with responsibility disclaimed. A CLI run from a terminal
  keeps the terminal's attribution.
- Use `daemon bootstrap` instead of `restart`, so a running updater is
  replaced too. An existing updater may still be attributed to an older
  Wizard, and it keeps proxy variables that were removed from .env.
  Preserve remote control, which bootstrap would otherwise reset.
- Skip the restart unless `daemon version` reports a managed daemon.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant