Add DXC root-cause investigation guidance

[Damyan Pepper (damyanp)]

Add a root AGENTS.md with concise guidance for investigating DXC compile-time regressions and miscompiles.

• Check local fork history and the LLVM 3.7 baseline before designing new mechanisms.
• Enumerate all consumers of shared analysis limits and heuristics.
• Account for the disabled upstream LLVM analysis and transform test suites.
• Validate that reproducers preserve the original failure's structural sensitivity.
• Use Increase scan limit for DSE, add option #2725 and Restore LLVM 3.7 MemoryDependenceAnalysis default #9003 as a short example of finding and restoring a fork-specific tuning regression.

This is contributor and agent guidance only, with no user-visible compiler behavior change, so it does not require a release note under CONTRIBUTING.md.

Assisted by GitHub Copilot.

[Copilot]

Copilot review overview

🟢 Approval recommended

The documentation-only change is narrowly scoped, with no unresolved findings or blocking issues.

Review effort: Balanced
Findings: None

What changed in this PR

Adds a selectively loaded DXC investigation skill to guide root-cause analysis without expanding repository-wide instructions.

Changes:

• Adds guidance on local fork history, shared tuning constants, and reproducer validation.
• Documents a memory-dependence regression example and test-coverage limitations.
• Clarifies why this documentation-only change needs no release note.

File	Description
.github/​skills/​root-cause-investigation/​SKILL.md	Defines the investigation skill, checklist, and worked example.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Copilot review overview

🟢 Approval recommended

The documentation is consistent with the supplied repository context and linked changes, with no unresolved issues.

Review effort: Balanced
Findings: None

[Chris B (llvm-beanz)]

A few comments, but otherwise this looks good.

[Copilot]

Copilot review overview

🟢 Approval recommended

This documentation-only change has no identified blocking issues and does not require a release note.

Review effort: Balanced
Findings: None

[joaosaffran]

LGTM, I personally would like to see some output examples produced by an agent following this rules, it would simplify evaluating if those make sense or not. Not sure if this is possible though

[joaosaffran]

Was this example produced by an agent following this rules? It would be nice to see some examples somewhere of how those rules are being applied, but maybe this is not possible... yet

[Damyan Pepper (damyanp)]

This example came about to try and improve the bad job an agent originally did trying to address this problem.