qinghu-sales-video
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for downloading Node.js binaries and associated checksum files from nodejs.org, which is a well-known and trusted technology service. It also facilitates the installation of the vendor-owned @iqinghu/qhkit package from the NPM registry.
- [COMMAND_EXECUTION]: The installation workflow includes a security best practice by instructing the agent to verify the downloaded Node.js binary using sha256sum before extraction, preventing the execution of corrupted or tampered files.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection through data ingestion.
- Ingestion points: Processes user-provided text prompts and local images/video files via the qhkit CLI as seen in SKILL.md.
- Boundary markers: Example commands do not utilize explicit delimiters to isolate untrusted user input within the JSON arguments.
- Capability inventory: Performs sub-process execution of the qhkit tool for video generation, task status polling, and estimation.
- Sanitization: Instructions do not specify pre-processing or escaping of the user-supplied strings before they are interpolated into the shell command.
Audit Metadata