code-search
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is to guide the agent in performing web searches for coding documentation and snippets. No malicious commands, obfuscation, or credential access were detected. The parameters and instructions provided are consistent with the stated purpose of finding reliable code references.
- [PROMPT_INJECTION]: The skill's use of web retrieval tools creates an inherent surface for indirect prompt injection from untrusted external content.
- Ingestion points: Web fetch and research tools (
web_fetch_exa,web_research_exa,web_search_exa, andweb_answer_exa) described inSKILL.md. - Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore instructions that might be embedded in the fetched documentation.
- Capability inventory: Web searching and data retrieval.
- Sanitization: No explicit instructions are provided for sanitizing or validating the content retrieved from external URLs before processing it.
Audit Metadata