vibe-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill requires including offending code lines verbatim in reports (e.g., showing hardcoded passwords/keys/tokens as "问题代码") and thus would force the LLM to output any secrets found in the repo, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly states it accepts "PR/MR URL → 通过平台工具获取diff" in SKILL.md, meaning it will fetch and read user-generated pull/merge request content from public hosting platforms and use that content to drive code-review decisions, so it clearly ingests untrusted third-party content that can influence actions.