exa
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from the open web through search results, page extraction, and research tasks.
- Ingestion points: Web content fetched via
/search,/contents, and/answerendpoints inSKILL.md. - Boundary markers: The instructions do not define clear delimiters or headers to isolate fetched web content from the agent's core instructions.
- Capability inventory: The skill allows the agent to read external data and synthesize it into answers, though it does not demonstrate higher-privilege capabilities like file system writes or arbitrary shell execution.
- Sanitization: While content moderation is an optional parameter, there is no explicit instruction to sanitize or escape the content of extracted web pages before processing.
Audit Metadata