release: 0.22.1

[seratch]

Release readiness review (v0.22.0 -> TARGET 4826f77)

This is a release readiness report done by $final-release-review skill.

Diff

v0.22.0...4826f77

Release intent

• Review mode: final candidate
• Intended release: patch 0.22.1
• Minimum required release type: patch
• Versioning verdict: compatible

Release call

🟢 GREEN LIGHT TO SHIP No confirmed release blocker remains. The candidate contains current main plus one three-file release commit, and its version metadata and frozen API contract agree.

Scope summary

• 343 files changed (+30573/-12664), covering runtime fixes, compatible API additions, tests, documentation, translations, and release tooling.
• The release commit changes only pyproject.toml, uv.lock, and tests/fixtures/released_api_contract.json.
• Python support and runtime dependency bounds remain unchanged.

Risk assessment (ordered by impact)

1. Encrypted Session transformations and compaction ownership are preserved

   • Risk: 🟢 LOW. The plaintext-storage regression found in the earlier candidate is resolved by fix(sessions): preserve Session transformations during generation tracking #4917.
   • Evidence: The identical public Runner probe now stores encrypted input and output in both normal and streamed runs, matching v0.22.0. Generation tracking calls the outer Session methods; encrypted compaction preserves decryption, TTL filtering, and original rollback tokens.
   • Files: src/agents/run_internal/session_persistence.py, src/agents/memory/openai_responses_compaction_session.py, src/agents/extensions/memory/encrypt_session.py.
   • Action: Preserve the wrapper boundary; direct underlying-backend mutations remain outside compaction concurrency guarantees.

2. Resume recovery and resource cleanup have explicit boundaries

   • Risk: 🟢 LOW. Failed resumed writes can recover without repeating completed work, while accepted terminal outputs fail closed when persistence fails.
   • Evidence: Pending writes and terminal-state markers survive RunState serialization. The new reader retains supported released schemas. Runner-created providers are closed without taking ownership of explicitly supplied providers; Voice and PTY cleanup preserve completion ordering.
   • Files: src/agents/run_state.py, src/agents/run.py, src/agents/result.py, src/agents/voice/, src/agents/sandbox/session/.
   • Action: Recover with the original Session and exclusive history access. Start a new run for an unrecoverable terminal state, and consume streams through their completion boundary.

3. Public additions preserve existing defaults and call shapes

   • Risk: 🟢 LOW. New options extend existing pipelines without requiring a minor release.
   • Evidence: MCP tool guardrails, blocked-output formatters, Voice settings/context, web-search image options, Docker labels, and Unix-local environment filtering are optional. Released positional ordering is preserved, and the contract freezes intended exports and concrete Voice classes.
   • Files: src/agents/run_config.py, src/agents/mcp/, src/agents/tool.py, src/agents/voice/, src/agents/sandbox/sandboxes/.
   • Action: Keep provider-specific qualifiers explicit. Unix-local environment filtering is opt-in and is not a stronger process-isolation boundary.

4. Publishing now requires explicit maintainer release operations

   • Risk: 🟢 LOW. Merging the release PR no longer creates a release tag automatically.
   • Evidence: Publishing validates tag provenance and package version and uses a protected PyPI deployment.
   • Files: .github/RELEASING.md, .github/workflows/publish.yml.
   • Action: Follow the documented manual tag, GitHub Release, and deployment-approval sequence after merge.

Documentation coverage (non-blocking)

• Coverage source: docs: document v0.22.1 behavior updates #4577 at head 4f569ed5c69a64cc25f9466af25ca0ed8c528138; all 16 current file patches reviewed.
• Status: partially covered.
• Covered obligations: provider ownership, Voice configuration/redaction, MCP guardrails, blocked-output formatting, resume recovery, compaction concurrency, image search, sandbox environment policy, Docker labels, tracing, and usage.
• Gaps or post-release suggestions:
  • docs/models/index.md: clarify paired Chat Completions tool outputs and the OpenAI-adapter scope of empty-truncation errors.
  • docs/tools.md: explain rejection of **kwargs keys colliding with named parameters.
  • docs/voice/pipeline.md and docs/sandbox/clients.md: explain cancellation consumption and final PTY-output settlement.
  • docs/sessions/encrypted_session.md: document encrypted compaction composition and its TTL/rollback boundary.
• Publication timing: publish guidance for unreleased behavior after 0.22.1 is available.

Notes

• No live provider requests were made during this review.
• The final candidate parent matches the last refreshed origin/main.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-08T09:04:47.646616Z	4826f77	Manual request
🔒 Security Review	✅ Completed 2026-09-08T09:06:31.235599Z	4826f77	Manual request

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 9b14fff3dd

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: cd7cce0e6b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 19f0378daf

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 70b2af3041

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[seratch]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Keep them coming!

Reviewed commit: 4826f77ab6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Security review completed. No security issues were found in this pull request.

Reviewed commit: 4826f77ab6

View security finding report

Only the user who started this review can view the report in Codex.

ℹ️ About Codex security reviews in GitHub

This is an experimental Codex feature. Security reviews are triggered when:

• You comment "@codex security review"
• A regular code review gets triggered (for example, "@codex review" or when a PR is opened), and you’re opted in so security review runs alongside code review

Once complete, Codex will leave suggestions, or a comment if no findings are found.