authority
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified in the analyzed skill files.
- [PROMPT_INJECTION]: The instructions focus purely on behavioral analysis and do not contain directives to bypass safety filters, ignore system prompts, or extract internal instructions.
- [DATA_EXFILTRATION]: There are no network requests, file system access commands, or mechanisms to export user data. No hardcoded credentials or sensitive paths were found.
- [REMOTE_CODE_EXECUTION]: The skill does not include scripts, package installations, or remote resource fetching. It is entirely prompt-based.
- [OBFUSCATION]: All content is in plain text (Chinese and English). No Base64, zero-width characters, homoglyphs, or encoded URLs were detected.
- [DYNAMIC_CONTEXT_INJECTION]: No dynamic shell command execution patterns (!
command) were found in the SKILL.md file.
Audit Metadata