google-web-search
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from the internet via Google Search, which represents a potential surface for indirect prompt injection where malicious instructions could be embedded in web content.
- Ingestion points: Content returned by the
google_web_searchtool (SKILL.md). - Boundary markers: The instructions provide citation rules but do not define explicit delimiters or warnings to the agent to disregard instructions within the fetched search content.
- Capability inventory: The skill describes search functionality; no scripts or tools for file writing, system execution, or data exfiltration are present in the provided skill definition.
- Sanitization: No filtering or sanitization of search snippets or image metadata is defined.
- [NO_CODE]: The skill consists entirely of instructional markdown and configuration metadata; no associated script files (Python, JavaScript, shell) were provided for analysis.
Audit Metadata