deep-research
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides a framework for the agent to systematically fetch and read external web content. This functionality creates an attack surface for indirect prompt injection where malicious instructions on external websites could attempt to influence the agent's subsequent actions, though this is the primary intended use case of the skill.\n
- Ingestion points: The skill utilizes
web_fetchto retrieve full content from external web pages as part of its Phase 2 and Search Strategy (SKILL.md).\n - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the fetched data.\n
- Capability inventory: The skill enables search operations and retrieval of external content via web tools.\n
- Sanitization: No specific filtering or sanitization steps are mandated for the external data before it is synthesized into output.
Audit Metadata