linkpix-kling-3-clone
Fail
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONPERSISTENCECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides instructions to download the Node.js runtime from
nodejs.orgor a well-known mirror (npmmirror.com). It includes a security best practice by piping the SHA256 checksum file togrepandsha256sum -cto verify the integrity of the downloaded archive before extraction. - [PERSISTENCE]: To ensure the installed Node.js binary is available in future sessions, the instructions suggest appending a path export command to the user's shell profile (
~/.bashrc). - [COMMAND_EXECUTION]: The skill executes various system commands, including
npm install -gfor the vendor's package@iqinghu/qhkit. It also usespython -candnpxto dynamically process and compress images if they exceed size limits. - [INDIRECT_PROMPT_INJECTION]: The skill accepts external video URLs (Douyin, TikTok, etc.) as input. The agent uses these links to reverse-engineer scripts which are then fed into AI generation models. This processing of untrusted external content represents a vulnerability surface where malicious metadata or script structures could influence the agent's behavior.
Recommendations
- HIGH: Downloads and executes remote code from: https://nodejs.org/dist/v22.22.3/SHASUMS256.txt - DO NOT USE without thorough review
Audit Metadata