researchers-financial
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external websites, which constitutes an attack surface for indirect prompt injection.\n
- Ingestion points: Data is retrieved from external financial sites and databases using
WebFetchandWebSearch(SKILL.md, SEC_FILING_GUIDE.md).\n - Boundary markers: The instructions do not include delimiters or specific warnings to differentiate untrusted external data from the agent's core instructions.\n
- Capability inventory: The skill is granted access to tools such as
Read,Edit,Write,Grep,Glob,WebFetch, andWebSearch(SKILL.md).\n - Sanitization: There is no evidence of data sanitization or filtering before processing external content.\n- [EXTERNAL_DOWNLOADS]: The skill fetches financial data and reporting from reputable organizations and well-known services, such as SEC.gov, Reuters, and Bloomberg.
Audit Metadata