Fix AI E2E tests for blocked XML-RPC endpoints

[iangmaia]

Description

Check the XML-RPC endpoint contract, require the Core REST editor path, and use a request-time REST policy to prevent mutation bypasses.
Also pins simulator-llm-pilot to a trunk release.

[dangermattic]

	1 Message
📖	This PR is still a Draft: some checks will be skipped.

Generated by 🚫 Danger

[crazytonyli]

Thanks for looking into the issue, Ian! The E2E test actually caught a bug, which is fixed in trunk. We should not need to do anything about the test suite now. I expect today's run to pass.

[Copilot]

Pull request overview

This PR updates the AI-driven iOS simulator E2E test harness to better handle test sites where XML-RPC is blocked (or intercepted by an edge 429), ensuring the app uses the Core REST editor path and preventing REST-based “mutation bypass” of UI failures.

Changes:

• Add an XML-RPC availability preflight (with an explicit “edge 429” allowance) and a small contract-test script to lock in the detector behavior.
• Pin simulator-llm-pilot installation to a specific upstream revision and add a runtime check for required CLI support (--rest-api-policy).
• Update AgentTests markdown scenarios and agent instructions to reflect the blocked XML-RPC contract and stricter REST boundaries.

Reviewed changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
Tests/AgentTests/ui-tests/text-post-publish.md	Updates publish assertions to require success without XML-RPC errors and not depend on a post-publish confirmation screen.
Tests/AgentTests/ui-tests/post-with-category-and-tag.md	Aligns prerequisites/cleanup with blocked XML-RPC and removes reliance on a specific category name.
Tests/AgentTests/ui-tests/create-scheduled-post.md	Adds blocked XML-RPC prerequisite and clarifies success criteria without extra confirmation UI.
Tests/AgentTests/ui-tests/create-blank-page.md	Adds blocked XML-RPC prerequisite and clarifies publish success criteria.
Tests/AgentTests/test-ai-e2e-contracts.sh	New bash contract tests for XML-RPC detection/preflight helpers.
Tests/AgentTests/app-instructions.md	Adds REST API usage boundaries (read-only verification; DELETE-only cleanup).
Scripts/ci/install-simulator-llm-pilot.sh	Fetches and installs simulator-llm-pilot from a pinned ref (or local override).
.buildkite/nightly.yml	Configures nightly AI E2E runs to accept an edge 429 from the XML-RPC endpoint.
.buildkite/commands/run-ai-e2e-tests.sh	Implements XML-RPC preflight, runs contract tests, enforces REST API policy, and improves script structure via main().

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[wpmobilebot]

📲 You can test the changes from this Pull Request in WordPress by scanning the QR code below to install the corresponding build.

	App Name	WordPress
	Configuration	Release-Alpha
	Build Number	33478
	Version	PR #25839
	Bundle ID	org.wordpress.alpha
	Commit	783c038
	Installation URL	1bj1oj48pvo2g

Automatticians: You can use our internal self-serve MC tool to give yourself access to those builds if needed.

[wpmobilebot]

📲 You can test the changes from this Pull Request in Jetpack by scanning the QR code below to install the corresponding build.

	App Name	Jetpack
	Configuration	Release-Alpha
	Build Number	33478
	Version	PR #25839
	Bundle ID	com.jetpack.alpha
	Commit	783c038
	Installation URL	2u6dl5h426jn8

Automatticians: You can use our internal self-serve MC tool to give yourself access to those builds if needed.

[mokagio]

@crazytonyli

Thanks for looking into the issue, Ian! The E2E test actually caught a bug, which is fixed in trunk. We should not need to do anything about the test suite now. I expect today's run to pass.

Does this mean your recommendation is to close this PR? Was the blocked XML-RPC endpoint something that the app should have been able to handle internally (hence the UI test failure fix) instead of something that we should handle in the tests scaffolding?