de-AI-writing
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust 'Security Boundary' (安全边界) in
SKILL.md, which explicitly instructs the agent to treat all user-provided text, files, and links as data rather than instructions. This is a best-practice mitigation against prompt injection and data manipulation attacks. - [SAFE]: All referenced files (
references/*.md) are local to the skill package. No external network requests, remote script downloads, or third-party package installations were detected, eliminating remote code execution and network exfiltration vectors. - [SAFE]: The skill's tools are limited to standard file operations (
Read,Write,Edit) and user interaction (AskUserQuestion). These tools are used within the scope of the skill's stated purpose of document editing and polishing. - [SAFE]: No obfuscation, hidden instructions, or credential exposure were found. The instructions are transparent and follow a logical structure for linguistic analysis.
Audit Metadata