dbs-action
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of markdown documentation and YAML configuration. No executable scripts (.sh, .py, .js) or binary files are present, eliminating the risk of direct code execution.
- [SAFE]: No network activity or data exfiltration. The instructions do not use tools like curl or wget, and they do not attempt to send user data to external servers.
- [SAFE]: Safety-first design. The skill includes a dedicated safety screening module (
references/safety-screening.md) that instructs the AI to recognize signs of psychological crisis or self-harm and immediately stop the diagnostic process to recommend professional help. - [SAFE]: No obfuscation or hidden logic. All instructions are provided in clear Chinese and English, with no use of Base64, zero-width characters, or homoglyph-based attacks to hide malicious intent.
- [SAFE]: Capability limitation. The skill does not request or use any high-privilege tools or sensitive file access (e.g., SSH keys, credentials).
Audit Metadata