Skip to content

fix(remote): correct fill recording recovery guidance - #4351

Merged
thymikee merged 2 commits into
callstack:mainfrom
nikitabhagat2307-prog:fix/remote-fill-record-as-guidance
Nov 10, 2026
Merged

thymikee merged 2 commits into
callstack:mainfrom
nikitabhagat2307-prog:fix/remote-fill-record-as-guidance

Conversation

@nikitabhagat2307-prog

@nikitabhagat2307-prog nikitabhagat2307-prog commented Nov 10, 2026 •

Copy link
Copy Markdown
Contributor

Summary

When fill --record-as lacks an armed recording, the daemon tags the error with a typed reason. Remote clients now explain that recording must start in a fresh session on a local daemon, instead of suggesting a command the remote transport rejects. Local recovery guidance and the intentional remote refusal remain unchanged. Adds focused daemon and client coverage. Four files changed; scope stays within fill validation and client error adaptation. Closes #4345.

Validation

  • Head: abdf3fa705b14f20fe6a1fd9014837f9cfffa0c9.
  • Build passed on the original implementation commit; focused Vitest passed on this head (2 files, 6 tests); oxlint passed on this head.
  • The affected gate on this head could not acquire its validation lock (EROFS at ~/.cache/agent-device/validation.reclaim.lock). On the previous commit it reached 76 Vitest files: 73 passed, 3 failed (333 passed, 6 failed), mostly because the environment could not create ~/.agent-device/apple-runner/leases; one proxy parity assertion also failed. Earlier gate receipt: .tmp/validation/run-7f69Hp/receipt.json.
  • GitHub checks remain required. Not merge-ready until the affected gate and remaining CI obligations are resolved. No device behavior changed.

@nikitabhagat2307-prog
nikitabhagat2307-prog marked this pull request as ready for review November 10, 2026 09:18

@cubic-dev-ai cubic-dev-ai Bot left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

All reported issues were addressed across 4 files

Reply to a comment to ask cubic a question or push back. It learns from your replies.

View guided diff | Re-trigger cubic

@nikitabhagat2307-prog
nikitabhagat2307-prog marked this pull request as draft November 10, 2026 09:32
@nikitabhagat2307-prog
nikitabhagat2307-prog marked this pull request as ready for review November 10, 2026 09:35

@cubic-dev-ai cubic-dev-ai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No issues found across 4 files

View guided diff | Re-trigger cubic

@nikitabhagat2307-prog

Copy link
Copy Markdown
Contributor Author

The review comments are addressed, and the latest automated review found no issues. Could a maintainer review this PR? The local affected-check gate was blocked by my environment; please let me know if you need any additional validation.

@thymikee

Copy link
Copy Markdown
Member

Thanks, this looks good at abdf3fa. Remote clients now get guidance they can follow: start a fresh session with open --save-script on a local daemon, keyed on the typed reason rather than the message text. There are no conflicts.

The comments on the remote text and the daemon test are fixed and can be resolved.

Not blocking: the reason string fill_record_as_requires_armed_recording is a bare literal in both the daemon and the client, so a rename on one side would silently turn off the hint without failing a test. One shared exported constant, like the other cross-package reasons, would close that.

@thymikee thymikee added the ready-for-human Valid work that needs human implementation, judgment, or maintainer merge label Nov 10, 2026
@thymikee
thymikee merged commit ba51a0e into callstack:main Nov 10, 2026
14 of 17 checks passed
@nikitabhagat2307-prog

Copy link
Copy Markdown
Contributor Author

Good point. The reason string is duplicated across the daemon and client, so a rename could disable the recovery hint without a test catching it. Since this PR is already merged, I will keep that separate from its completed fix and address it in a follow-up if the maintainers want it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready-for-human Valid work that needs human implementation, judgment, or maintainer merge

Projects

None yet

Development

Successfully merging this pull request may close these issues.

fill --record-as recovery recommends --save-script although remote-client mode rejects it

2 participants