company-news-research
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill outlines a manual or agentic research workflow using standard data sources and templates.
- [PROMPT_INJECTION]: The skill is designed to ingest data from untrusted external sources including news outlets, social media, and forums. Ingestion points: Workflow steps 2 and 3 in SKILL.md and the source priority list in news-source-policy.md. Boundary markers: Explicit instructions are provided in the Quality Gate to distinguish rumors and framing from primary facts, and to use the provided News Digest Template. Capability inventory: The skill performs file writes to the 'research/targets//evidence/news/' directory to save news digests. Sanitization: No explicit sanitization or escaping of external content before interpolation or file writing is documented.
Audit Metadata