content-extractor
Warn
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
subprocess.runto execute thecurlsystem command for fetching web content and downloading media. This is used in the Xiaoyuzhou platform's 'fast mode' extraction logic. - [REMOTE_CODE_EXECUTION]: The
_extract_wechatmethod incontent_extractor.pyattempts to import a module from a hardcoded absolute local filesystem path (/Users/delta/.openclaw/workspace/skills/wechat-article-fetcher), which constitutes dynamic loading from a environment-specific path. - [EXTERNAL_DOWNLOADS]: The skill performs network requests and downloads media files from various third-party platforms including
xiaoyuzhoufm.com,douyin.com, andbilibili.com. - [PROMPT_INJECTION]: The skill ingests untrusted content from social media platforms and returns it without boundary markers or sanitization. This creates a surface for indirect prompt injection if the agent processes the extracted text as instructions.
Audit Metadata