case-control-study-planner
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional, providing a structured approach for medical researchers to plan study protocols. No malicious code, dangerous shell commands, or security vulnerabilities were detected across the skill files and reference modules.
- [PROMPT_INJECTION]: The instructions and integrated reference modules do not contain any patterns intended to bypass AI safety filters, ignore system instructions, or extract system prompts. The expert persona is used appropriately for its intended clinical research context.
- [DATA_EXFILTRATION]: There is no evidence of data exfiltration or unauthorized data exposure. The skill does not access sensitive environment variables, credentials, or local configuration files, and it performs no network operations for data transmission.
- [REMOTE_CODE_EXECUTION]: No remote code execution patterns or dynamic code generation techniques were found. The skill does not download or execute external scripts or packages.
- [EXTERNAL_DOWNLOADS]: The skill contains a reference to the author's GitHub repository for documentation purposes, which is consistent with its stated purpose and does not involve the execution of untrusted remote content.
Audit Metadata