web-research
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves processing untrusted data from the web, which constitutes an attack surface for indirect prompt injection.
- Ingestion points: Content retrieved from external websites using the
web_fetchtool (SKILL.md). - Boundary markers: The instructions do not mandate the use of delimiters or warnings to ignore instructions that may be embedded in retrieved web content.
- Capability inventory: The agent is authorized to use
web_searchandweb_fetchto retrieve and synthesize data into reports. - Sanitization: No explicit procedures are defined for sanitizing or filtering the content fetched from the web before processing.
Audit Metadata