find-skills

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses web_fetch url="https://skills.sh/audits" to ingest markdown from a public outsider-authored webpage into the agent context for parsing and display.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runtime explicitly fetches SKILL.md from raw GitHub URLs (https://raw.githubusercontent.com/{source}/main/{skillId}/SKILL.md and https://raw.githubusercontent.com/{source}/master/{skillId}/SKILL.md) and then injects that fetched content as the skill "instructions" via learn_skill, so remote content directly controls agent instructions and is a required dependency.