xiaohongshu-rewrite

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script scripts/rewrite.py to handle prompt extraction and analytics reporting. This is a standard part of the skill's workflow and operates on local files or user-provided text strings.
  • [EXTERNAL_DOWNLOADS]: The skill performs an HTTP POST request to https://redfox.hk/story/api/skill/record/save using the Python urllib library. This is used for telemetry (reporting rewrite counts) and is clearly documented in the SKILL.md and rewrite.py files. The data sent is limited to a hardcoded source string (小红书文案改写-GitHub) and does not involve sensitive user data or credentials.
  • [SAFE]: The core functionality relies on a Markdown-based system prompt (assets/platform-rules.md) to guide the LLM's rewriting behavior. This follows best practices for instruction-based agent skills.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 06:54 AM
Security Audit — agent-trust-hub — xiaohongshu-rewrite