[Krrish Mittal (takyyon)]

Commit Type

• feature - New functionality
• fix - Bug fix
• refactor - Code restructuring without behavior change
• perf - Performance improvement
• docs - Documentation update
• test - Test-related changes
• chore - Maintenance/tooling

Risk Level

• Low - Minor changes, limited scope
• Medium - Moderate changes, some user impact
• High - Major changes, significant user/system impact

What & Why

PR AI Validation fails before assessment when GitHub rejects a raw diff over 20,000 lines. This is the required-check failure observed on #9677.

Fall back to bounded, paginated native file patches only for the structured HTTP 406 PullRequest/diff/too_large error. Preserve normal raw-diff retrieval, explicitly identify unavailable patches, and propagate unrelated errors.

Keep the existing 4,000-line gather bound and 2,000-line/100,000-byte prompt limits. Apply the line and byte limits through an intermediate temporary file instead of a producer pipe: large inputs otherwise cause SIGPIPE under Actions' strict Bash settings.

Run actual-workflow regression tests from the existing trusted checkout. Permissions, models, verdict handling and the pull_request_target trust boundary remain unchanged; no PR-head checkout is introduced.

Impact of Change

• Users: No product UX or API changes.
• Developers: Oversized PRs can reach assessment without weakening validation or silently accepting missing input.
• System: Changes only the validator workflow and its regression test script. Because the workflow runs from the trusted base, this fix must be integrated there before a fresh PR event can use it.

Test Plan

• Unit tests added/updated

• E2E tests added/updated

• Manual testing completed

• Tested in: local focused workflow/test checks.

• node --test scripts/pr-ai-validation.test.mjs: 30 passed, 0 failed.

• Tests execute the actual extracted workflow scripts, including strict Bash with two-megabyte inputs, pagination bounds, missing patches, exact line/byte limits, input/temp-file failures and trusted-checkout test wiring.

Contributors

Krrish Mittal (@takyyon)

Screenshots/Videos

N/A (CI workflow/test-only change).

[github-actions]

🤖 AI PR Validation Report

PR Review Results

Thank you for your submission! Here's detailed feedback on your PR title and body compliance:

✅ PR Title

• Current: fix(ci): handle oversized PR diff fallback safely
• Issue: The title is specific, follows a conventional format, and accurately describes the CI behavior change.
• Recommendation: No changes required.

---

✅ Commit Type

• Properly selected (fix).
• Only one commit type is selected, and the workflow bug fix is consistent with this classification.

---

✅ Risk Level

• The Medium risk level is selected in the body and matches the risk:medium label.
• The advised risk level is medium, based on changes to a pull_request_target validation workflow, GitHub API error handling, pagination, and prompt-input processing. The documented trust boundary is preserved, so a high-risk classification is not required.

---

✅ What & Why

• Current: Clearly explains the oversized-diff failure, the structured fallback behavior, sampling limits, temporary-file change, testing approach, and preserved security assumptions.
• Issue: No compliance issues identified.
• Recommendation: No changes required.

---

✅ Impact of Change

• The impact section identifies the absence of product UX/API changes and explains the effects on developers and the validation workflow.
• Recommendation:
  • Users: The stated absence of product UX or API changes is appropriate.
  • Developers: The explanation of oversized PR handling and validation behavior is sufficient.
  • System: The workflow scope, trusted checkout behavior, and unchanged trust boundary are documented appropriately.

---

✅ Test Plan

• Unit tests are marked and a test file is added in the diff (scripts/pr-ai-validation.test.mjs). The described 30 passing tests cover normal diff retrieval, structured oversized-diff fallback, pagination limits, unavailable patches, error propagation, strict Bash sampling, and trusted-checkout wiring.
• Manual testing is also documented with the command and result. The test plan passes because unit tests are present in the diff.

---

⚠️ Contributors

• @takyyon is credited.
• Consider adding PMs, designers, reviewers, or other contributors if they helped shape or validate this change; this is optional and does not block approval.

---

✅ Screenshots/Videos

• Correctly marked as not applicable because this is a CI workflow and test-only change with no visual product changes.

---

Summary Table

Section	Status	Recommendation
Title	✅	No changes required.
Commit Type	✅	fix is correctly selected.
Risk Level	✅	Medium matches both the body and risk:medium label.
What & Why	✅	Explanation is clear and sufficiently detailed.
Impact of Change	✅	User, developer, and system impacts are covered.
Test Plan	✅	Unit tests are present in the diff and manual testing is documented.
Contributors	⚠️	Add other contributors if applicable.
Screenshots/Videos	✅	Correctly identified as not applicable.

---

The PR title and body pass review. No required updates are needed.

---

Powered by: Logic App (fallback) | Last updated: Fri, 30 Oct 2026 16:45:50 GMT

[Copilot]

🟢 Approval recommended

No active review issues remain for the oversized-pull-request validation changes.

No open issues 🎉

What changed in this PR

This PR repairs AI PR Validation for oversized pull requests while preserving normal diff handling and existing validation policy.

Changes:

• Falls back to bounded, paginated file patches only for GitHub’s structured oversized-diff error; other errors still propagate.
• Separates shell sampling reads to avoid SIGPIPE failures while retaining line and byte limits.
• Adds regression tests and runs them from the trusted checkout before gathering PR data.

File	Description
scripts/​pr-ai-validation.test.mjs	Tests fallback handling, sampling limits, failure propagation, and trusted-checkout configuration.
.github/​workflows/​pr-ai-validation.yml	Adds the guarded fallback, fixes shell sampling, and integrates regression tests.

_{Review effort: Balanced}