Test fixtures inherit the Codex Desktop marker and select codex-app over fake terminal drivers
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 76/100
Research direction
Read tests/test_helper.bash to see where fixtures clear ambient environment variables, then inspect scripts/drivers/terminals/codex-app/ops.sh and the native Codex App detection tests. Clear the desktop marker for ordinary fixtures and restore it explicitly in tests that exercise native detection. Run the affected Bats tests with and without the marker to confirm fake drivers are selected in fixtures while native detection still works when opted in.
Written by the indexing model from the issue text.
Description
Environment and reproduction
agmsg v1.6.0-rc.4, clean commit 3fb012a86203f227d67dc986b18c15733e702873, macOS, Codex App, Bats1.13.0. Tests use their existing fake terminal drivers and sandboxed HOME/TMP; no real terminal or message store was modified.
CODEX_INTERNAL_ORIGINATOR_OVERRIDE='Codex Desktop' bats tests/test_session_start_terminal_line.bats
env -u CODEX_INTERNAL_ORIGINATOR_OVERRIDE bats tests/test_session_start_terminal_line.bats
With the inherited marker, the plain/herdr/tmux cases choose codex-app; the same selected cases pass when only that variable is removed in the child environment. Equivalent baseline/control runs independently reproduce failures across self-fix, self-name, self-rename, terminal registry/context, capabilities, delivery, despawn, dispatch and spawn tests. The separate storage compatibility failures remain after marker removal.
Cause and expected boundary
tests/test_helper.bash clears ambient tmux, herdr, Claude profile and session markers at load/setup, but leaves CODEX_INTERNAL_ORIGINATOR_OVERRIDE. scripts/drivers/terminals/codex-app/ops.sh detects exact Codex Desktop; its priority1/exclusive detection then wins over fake terminal drivers installed by a test. This is intended production precedence applied to an unintentionally contaminated fixture.
Fixtures should clear the ambient desktop marker alongside the existing host/session variables. Tests intentionally exercising native Codex App detection should explicitly restore it. The native desktop-driver opt-in control passes under a clean child environment, so this report does not propose lowering production driver priority or disabling native delivery.
Child-only removal is a diagnostic workaround; no global App setting or production environment change is requested. This failure reproduces on unmodified RC and is separate from the Claude transcript-observer fix.
- Dominant language
- Shell
- Stars
- 1.5k
- Forks
- 156
- Avg merge
- 16h 23m
- Merged PRs (30d)
- 197
Getting set up
- No Dockerfile or Docker Compose file
- No pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from fujibee/agmsg
-
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 88/100
Maintainers usually reply within 1 day
-
[P2] app: peer-message previews pass C0/DEL control bytes to pty_injectPossibly taken A pull request linked to this issue is open or already merged. Open
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
Maintainers usually reply within 1 day
Similar issues
-
diagnosing-superpowers: transcript-as-evidence guard never reaches the dispatcher or the analystsOpen
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
obra/superpowers#2662 ·
Maintainers usually reply within 3 days
-
package-update
Difficulty 1/5 Under an hour Newbie friendliness 69/100
oSoWoSo/oco#844 · 2 comments ·
Maintainers usually reply within 1 day
-
bug linux priority-low
Difficulty 2/5 1-3 hours Newbie friendliness 73/100
Maintainers usually reply within 1 day
-
chart-check misses a config variable whose env call spans linesPossibly taken @thebadalov claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
kubemg/kubemg#291 · 2 comments ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
rossoctl/automation#130 ·
Maintainers usually reply within 3 days