code-pr-review

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The review explicitly requires detecting and reporting "hardcoded secrets, API keys, passwords" with file/line references and optional code snippets while ingesting raw diffs or PRs, which will lead the model to reproduce secret values verbatim if they appear in the input.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly states it will fetch a GitHub/GitLab PR URL "via web_fetch" at runtime (e.g., https://github.com///pull/123), and that fetched diff would be injected into the agent context to drive the review prompts, so external content directly controls the agent's instructions.