vhscli
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill uses
npxto download and execute the@getvhs/vhscli2package, which is the official client for the vendor's services. - [COMMAND_EXECUTION]: The skill interacts with the local environment by executing shell commands to run media generation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted user prompts and file paths.
- Ingestion points: User-provided text prompts and local media file paths used as arguments in shell commands.
- Boundary markers: The instructions explicitly suggest using stdin (
-) for prompts to avoid shell escape character vulnerabilities. - Capability inventory: Execution of shell commands through the
npxutility. - Sanitization: The skill recommends utilizing standard input for prompt handling, reducing the risk of shell-based injection.
Audit Metadata