test: validate ai-review.yml@main invocation end-to-end

[chouetz]

Purpose

Pure validation PR - confirms the post-merge state of PR #19 works correctly:

• `ai-review.yml` on `main` now points `uses:` at `@main` (flipped from the feature branch pointer now that main has `prompt_file_pattern` declared).
• `prompt_file_pattern: '**/codereview_guideline.md'` should discover `codereview_guideline.md` at the repo root via the self-checkout + `discoverGuidelinesForChangedFiles` path.
• `fibonacci.py` has an intentional off-by-one bug for the reviewer to catch, confirming the full pipeline end-to-end.

Will close without merging once validated.

🤖 Generated with Claude Code

[chouetz]

/dd-review

[github-actions]

AI review by Claude - workflow run

Review verdict: patch is incorrect

The fibonacci function has an off-by-one error in its loop bound. Starting from a, b = 0, 1, computing fib(n) requires n iterations, but the loop uses range(n - 1). This makes it return fib(n-1) for every n >= 1 — e.g. fibonacci(1) returns 0 instead of 1, and fibonacci(9) returns 21 instead of 34. The __main__ demo would print a sequence shifted by one position. See the inline comment for the fix.

Confidence: 0.97

[github-actions]

Off-by-one error in the loop bound. With the accumulator initialized to a, b = 0, 1, producing fib(n) needs n iterations, but range(n - 1) runs one too few. As written the function returns fib(n-1) for all n >= 1: fibonacci(1) yields 0 (should be 1), fibonacci(9) yields 21 (should be 34). The n == 0 early return already covers the zero case, so the loop count should simply be n.

Suggested change

	a, b = 0, 1
	for _ in range(n - 1):
	for _ in range(n):

[chouetz]

Validated: ai-review.yml@main invocation works end-to-end (self-checkout resolved to main correctly, prompt_file_pattern discovered codereview_guideline.md, review caught the intentional fibonacci bug). Closing without merging.