deep-research
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the web, which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: Web search results (SKILL.md, step 2).
- Boundary markers: None specified to isolate search data from system instructions.
- Capability inventory: Web search tool, Word document generation tool.
- Sanitization: The skill mandates cross-checking facts and citing sources but does not include explicit instructions to ignore embedded prompts in search results.
- [SAFE]: The skill demonstrates legitimate research behavior. No evidence of credential theft, remote code execution, obfuscation, or persistence mechanisms was found. It relies on standard platform tools for its core functionality.
Audit Metadata