market-analysis
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the web and local research files, creating a potential surface for indirect prompt injection.\n
- Ingestion points: Untrusted data enters the agent context through
WebSearchandWebFetchtool outputs, as well as existing research evidence files located intmp/research/.\n - Boundary markers: The skill lacks instructions to wrap external content in delimiters or to explicitly ignore embedded instructions within the research data, which could lead the agent to follow malicious commands hidden in web content.\n
- Capability inventory: The skill has significant capabilities, including the ability to write to the file system (
docs/and.claude/directories), perform network operations via research tools, and spawn subagents usingspawn_agent.\n - Sanitization: No sanitization or validation steps are included to clean external data before it is written to the final output artifacts.
Audit Metadata