component-patterns

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly spawns an external-api-researcher when the api_docs_url is provided and uses WebFetch/WebSearch to fetch and analyze external API documentation pages (see "Phase: API Research" in SKILL.md and references/api-research-guide.md), which means it ingests open/public third‑party URLs that can influence subsequent agent decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly spawns an external-api-researcher that WebFetches the user-provided api_docs_url at runtime and uses the fetched documentation to produce api-research.json which directly informs prompts/instructions and generated code (flagged URL: the user-supplied api_docs_url documentation URL).