pr-tour

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires copying PR diff content (oldPath/diffHeader/diffText/diffText) verbatim into the generated JSON/HTML, which would force the model to reproduce any secrets that appear in those diffs.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). Required workflow step 2–3 fetches an outsider-authored GitHub PR’s diff (and PR body) via gh pr view/gh pr diff, then the agent ingests that diff text to group hunks and write prose before rendering.