soc2
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill serves as a purely instructional and advisory resource for organizations preparing for SOC 2 audits. It provides high-quality guidance on Trust Services Criteria, gap analysis, and control documentation without requesting or performing dangerous operations.
- [NO_CODE]: The skill and its associated reference files consist entirely of Markdown documentation. There are no scripts, binaries, shell commands, or automated tasks that could be executed on a host system.
- [PROMPT_INJECTION]: The instructions are focused on providing compliance advice. No evidence of prompt injection, jailbreak attempts, or safety guideline bypasses was detected. The role-play instruction ("You are an expert SOC 2 compliance advisor") is standard for defining the AI's persona and context.
- [DATA_EXPOSURE]: No hardcoded secrets, API keys, or credentials are present. Mentions of sensitive file paths (e.g., AD logs, AWS console exports) occur strictly within the context of examples for audit evidence and do not instruct the agent to access local user files.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any network requests, download external scripts, or include third-party software dependencies.
Audit Metadata