market-overview
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Prompt Injection] (LOW): The skill is susceptible to indirect prompt injection via the processing of untrusted news data.
- Ingestion points: The variable
{news_stories}in SKILL.md accepts external data likely sourced from news feeds or research snippets. - Boundary markers: Absent. The prompt does not use delimiters or clear boundaries to isolate the
{news_stories}variable from the core instructions. - Capability inventory: No code execution, file system access, or network capabilities are defined in the provided skill. The risk is limited to output manipulation (e.g., inserting malicious links or misinformation into the generated overview).
- Sanitization: None. The instructions do not warn the model to ignore potential commands embedded within the news data.
Audit Metadata