public-health-analyst
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns such as prompt injection, credential harvesting, or unauthorized network operations were detected in the skill files.
- [EXTERNAL_DOWNLOADS]: The skill provides URLs to trusted public health resources, including the CDC (wonder.cdc.gov), Healthy People 2030 (health.gov), and professional organizations like APHA and CSTE. These are documented as reference sources for data analysis and are treated as safe according to standard security practices for well-known services.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external health data (e.g., surveillance reports, community surveys).
- Ingestion points: User-provided datasets or clinical reports mentioned in Phase 1 and Phase 2 of the workflow (SKILL.md).
- Boundary markers: The skill lacks explicit delimiters for user data, but provides structured decision frameworks to guide analysis.
- Capability inventory: No file-system writes, subprocess calls, or network-active tools are included in the skill definition, mitigating technical risks.
- Sanitization: None explicitly defined. The risk is assessed as safe/low as the persona lacks the tools to execute malicious payloads embedded in data.
Audit Metadata