find-skills

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

• Third-party content exposure detected (low risk: 0.10). The enrichment script fetches and ingests webpage text from user-published skill pages on skills.sh, where an outsider can publish arbitrary descriptions and content.

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

• Potentially malicious external URL detected (high risk: 0.70). The script fetches descriptions from arbitrary third-party web domains (skills.sh and agent-skills.md) at runtime and relies on their content for behavior selection without strong publisher provenance or integrity verification, representing a weak operational dependency.