Allow automated commits and fix C-API test instructions

[moreal]

Summary

I've cleaned up the commit instructions in AGENTS.md so that automated tasks using standard git commit—like scripts/update_lib quick—can generate local commits without bypassing pre-commit hooks.

CONTRIBUTING.md already specifies that rustpython-capi should be excluded from workspace tests and tested within the crates/capi directory to ensure separate Cargo settings are applied. However, the test commands in AGENTS.md were missing the rustpython-capi exclusion, causing an inconsistency between the two documents.

Running C-API tests from the repository root prevents crate-specific PyO3 settings from being applied, leading to a mix of CPython ABI and RustPython object layouts, which can cause SIGSEGV in C-API tests. Therefore, I've updated AGENTS.md to align with CONTRIBUTING.md and run C-API tests separately.

Finally, I've specified relevant links and formats to ensure that AI-assisted contributions and commits follow RustPython's AI Policy, including the use of the Assisted-by trailer in commit messages.

Summary by CodeRabbit

Documentation

• Updated contributor guidance for AI policy compliance and disclosure.
• Clarified commit-hook installation, execution, automatic fixes, restaging, and retry behavior.
• Expanded validation guidance to cover workspace, C-API, and change-specific tests.

[coderabbitai]

📝 Walkthrough

Walkthrough

AGENTS.md adds AI policy and commit disclosure requirements. It updates hook instructions to require installation, execution, auto-fix restaging, and commit retries. It also documents task-specific validation, including separate workspace and C-API test commands.

Changes

Contribution guidance

Layer / File(s)	Summary
Workflow and validation requirements AGENTS.md	The instructions add AI disclosure requirements, require non-bypassed hook execution with restaging and commit retries, and define conditional validation commands. The workspace test command excludes rustpython-capi, which has a separate test command.

Estimated code review effort: 1 (Trivial) | ~3 minutes

Possibly related PRs

• RustPython/RustPython#7914: Adds related AI policy and contribution requirements.
• RustPython/RustPython#7963: Updates related pre-commit and auto-fix guidance.
• RustPython/RustPython#8427: Updates related workspace and C-API test guidance.

Suggested labels: skip:ci

Suggested reviewers: shaharnaveh, leehanjeong, youknowone

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the two main changes: allowing automated commits and correcting C-API test instructions.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

AGENTS.md (1)

20-23: 🔒 Security & Privacy | 🔵 Trivial | ⚡ Quick win

Include the pull-request disclosure requirement.

The concrete disclosure bullet only covers commit messages. The linked AI Policy also requires disclosure in the pull-request description and a statement of the extent of AI assistance. Add those requirements here, or state that the linked policy is authoritative for the complete disclosure process. (github.com)

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@AGENTS.md` around lines 20 - 23, Update the AI Policy section in AGENTS.md to
explicitly require disclosure of AI assistance in pull-request descriptions,
including the extent of assistance, or clarify that the linked AI Policy governs
the complete disclosure process. Preserve the existing commit-message trailer
requirements.

Source: MCP tools

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@AGENTS.md`:
- Line 35: Clarify the testing guidance in AGENTS.md by rewriting the
interpreter test instruction so `<module>` is explicitly a placeholder for one
or more relevant test modules, without leaving “modules” as an apparent literal
command argument. Preserve the existing cargo command and conditions for changes
touching Lib/ or interpreter behavior.

---

Nitpick comments:
In `@AGENTS.md`:
- Around line 20-23: Update the AI Policy section in AGENTS.md to explicitly
require disclosure of AI assistance in pull-request descriptions, including the
extent of assistance, or clarify that the linked AI Policy governs the complete
disclosure process. Preserve the existing commit-message trailer requirements.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yml

Review profile: CHILL

Plan: Pro Plus

Run ID: 7a5daaa1-5eb1-4cd3-8093-d49a2de8bcfc

📥 Commits

Reviewing files that changed from the base of the PR and between 68bcb8c and be86d03.

📒 Files selected for processing (1)

• AGENTS.md

[youknowone]

Thank you!