Skip to content

Explain Codex task provider persistence - #33

Merged
theFong merged 1 commit into
theFong:mainfrom
swaminator:fix/codex-stale-task-provider
Aug 27, 2026
Merged

theFong merged 1 commit into
theFong:mainfrom
swaminator:fix/codex-stale-task-provider

Conversation

@swaminator

Copy link
Copy Markdown
Contributor

Summary

  • explain that Codex persists the model provider when each task is created
  • require a new task after installing or reconnecting before selecting a Webster model
  • print the guidance for managed, unmanaged, already-restarted, and stopped app-server states
  • document the ChatGPT-account error produced when an older OpenAI task selects a Webster model

Validation

  • bash -n install.sh codex-setup.sh test.sh

  • ./test.sh

  • npm test in codex-model-proxy

  • shellcheck codex-setup.sh

  • git diff --check

  • actionlint .github/workflows/bootstrap.yml

  • This change includes validation that would fail if the new behavior broke.

  • Relevant success and failure paths are tested.

  • Safe re-run/idempotency behavior is preserved.

  • The supported GitHub Actions matrix passes.

Documentation

  • User-facing documentation was updated, or no documentation change is needed.

@theFong
theFong merged commit 5384a91 into theFong:main Aug 27, 2026
7 checks passed
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.

2 participants