fact-check
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts external documents as input for analysis without implementing security boundaries, which creates an attack surface for indirect prompt injection.
- Ingestion points: The target document identified in Step 1 and processed in Steps 2 and 3 serves as the primary ingestion point for untrusted data.
- Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or block quotes) or explicit instructions to ignore embedded commands within the target document.
- Capability inventory: The skill leverages the
WebSearchtool and theTasktool (for sub-agent execution), which could be misused if a malicious document successfully influences the agent. - Sanitization: There is no evidence of input validation, escaping, or sanitization of the document content prior to model processing.
Audit Metadata