search-skill

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (Workflow steps 2–3 and the "SKILL.md Fetch Pattern" in reference.md) explicitly performs WebSearch/WebFetch against public sources like GitHub, SkillHub (skillhub.club) and the broad web to fetch and parse third‑party SKILL.md files, which are untrusted/user‑generated content that the agent reads and uses to drive recommendations and decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly performs a WebFetch at runtime to read SKILL.md from raw GitHub URLs (e.g. https://raw.githubusercontent.com/{owner}/{repo}/main/.claude/skills/{name}/SKILL.md), and that fetched content is parsed/evaluated to drive the agent's assessment and outputs, so remote files can directly control prompts.