test-guide-from-code

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

• Third-party content exposure detected (medium risk: 0.30). The skill workflow ingests code diffs and PR/MR content from external repositories via CLI commands and user input, which can contain outsider-authored free text and injection payloads.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill references an external repository (FuDesign2008/open-skills) via an npx skills add command in SKILL.md line 20, which is an operational setup dependency retrieved from an arbitrary user-owned GitHub repository.