Prevent zsh correction prompts from blocking agent commands

[bivo]

Use zsh-specific command decoration for agent-requested commands while preserving pager handling and existing execution guards.

Description

Fixes an Agent Mode regression in interactive zsh sessions with correction enabled.

When the agent runs a mistyped command (for example zef README.md), zsh can prompt:

zsh: correct 'zef' to 'zed' [nyae]?

That prompt blocks autonomous agent execution. This PR suppresses zsh correction only for agent-requested commands by decorating those commands with nocorrect, while preserving existing command execution behavior and pager handling.

Linked Issue

Closes #13613

Testing

Manually tested with ./script/run.
Reproduced baseline behavior in stable build:

zsh: correct 'zef' to 'zed' [nyae]?

appears and can block flow

Verified fixed behavior in local build:
◦ agent-requested mistyped command does not block on correction prompt
◦ command fails normally (e.g. command not found)
Verified manual user-typed commands still keep normal zsh correction behavior

Also auto-tests are updated.

Screenshots / Videos

How it works currently in production build:

And how it works after fix in this PR:

Agent Mode

• Warp Agent Mode - This PR was created via Warp's AI Agent Mode

[cla-bot]

Thank you for your pull request and welcome to our community. We require contributors to sign our Contributor License Agreement, and we don't seem to have the users @bivo on file. In order for us to review and merge your code, each contributor must visit https://cla.warp.dev to read and agree to our CLA. Once you have done so, please comment @cla-bot check to trigger another check.

[bivo]

@cla-bot check

[cla-bot]

The cla-bot has been summoned, and re-checked this pull request!

[oz-for-oss]

@bivo

Every PR must be linked to a same-repo issue before Oz can review it.

Next step: open or find a same-repo issue describing this change, then link it to this PR by adding Closes #123 to the PR description (or using the "Development" sidebar on GitHub). A maintainer will mark the issue ready-to-implement when it is ready. Once it is marked, comment /oz-review to re-trigger review.

See the contribution guidelines for the full readiness model.

Powered by Oz

[oz-for-oss]

@bivo

Every PR must be linked to a same-repo issue before Oz can review it.

Next step: open or find a same-repo issue describing this change, then link it to this PR by adding Closes #123 to the PR description (or using the "Development" sidebar on GitHub). A maintainer will mark the issue ready-to-implement when it is ready. Once it is marked, comment /oz-review to re-trigger review.

See the contribution guidelines for the full readiness model.

Powered by Oz