SafeAI-Global PRD Agent
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill suite is a highly professional and security-conscious toolset designed for regulatory compliance. It uses a modular knowledge engine to retrieve up-to-date legal data without relying on hardcoded logic.
- [PROMPT_INJECTION]: No malicious bypass or system-override instructions were detected. The skill uses standard instructional language to manage its multi-mode compliance scanning.
- [DATA_EXFILTRATION]: No unauthorized data access or network transmission patterns identified. The skill implements a 'Data Redaction Layer' that actively flags and masks personally identifiable information (PII) before outputting content.
- [REMOTE_CODE_EXECUTION]: The skill contains no executable code or remote script downloads. The author explicitly documented the removal of remote fetch strings to mitigate supply-chain risks.
- [CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or credentials found. The skill includes specialized tools to detect and remediate hardcoded secrets in user-submitted code snippets.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted user data (PRDs and source code), it defines clear boundary markers, requires user confirmation for sensitive data handling, and is focused on security auditing, making it a low-risk implementation of its primary purpose.
Audit Metadata