earnings-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and process external financial data from various websites, creating a surface for indirect prompt injection.
- Ingestion points: Financial documents including earnings releases, 10-Q filings, and transcripts are retrieved from external sites such as company investor relations pages and third-party financial platforms (e.g., Seeking Alpha) as described in the Phase 1 instructions in
workflow.md. - Boundary markers: The instructions do not specify the use of delimiters or protective text wrappers (e.g., 'ignore embedded instructions') when the agent processes the retrieved external data.
- Capability inventory: The agent is empowered to execute Python code (using
matplotlib,pandas, andseaborn) for chart generation and to create structured DOCX files. - Sanitization: The skill lacks instructions for sanitizing or filtering the content of the retrieved external documents before they are used to influence the agent's report generation logic.
Audit Metadata