douyin-video-extractor
Fail
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to download the 'uv' package manager via a piped shell script from 'astral.sh', which is the official and well-known domain for this tool. It also clones a project repository from GitHub ('yzfly/douyin-mcp-server') to provide its core functionality.
- [COMMAND_EXECUTION]: The installation and setup process involves executing several shell commands, including system package managers ('apt', 'brew', 'choco') to install FFmpeg, and Python-specific execution commands ('uv run', 'uvx') to start the server and CLI tools.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and transcribes untrusted audio content from Douyin/TikTok videos.
- Ingestion points: The 'extract_douyin_text' tool and 'VideoTranscriber' class process audio from external URLs.
- Boundary markers: The instructions do not define clear delimiters or specific instructions for the agent to ignore potentially malicious commands embedded within the transcribed text.
- Capability inventory: The skill possesses file-writing capabilities through 'transcriber.save_markdown' and command execution via 'uv run'.
- Sanitization: There is no evidence in the instructions of sanitization or filtering applied to the transcribed text before it is returned to the agent context.
Recommendations
- HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata