video-prompting-guide
Fail
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- REMOTE_CODE_EXECUTION (CRITICAL): The skill explicitly instructs the user to execute
curl -fsSL https://cli.inference.sh | sh. This pattern is a major security risk because it downloads a script from the internet and pipes it directly into a shell interpreter without allowing for inspection or verification. - EXTERNAL_DOWNLOADS (MEDIUM): The skill relies on binaries and scripts hosted at
inference.shanddist.inference.sh. These domains are not part of the trusted source list (e.g., official GitHub organizations or well-known package registries), increasing the risk of supply chain attacks. - COMMAND_EXECUTION (LOW): The skill requests permission for the
Bashtool to runinfsh *commands. While this is the intended functionality for the platform, the prerequisite installation step bypasses standard security controls. - INDIRECT_PROMPT_INJECTION (LOW): The skill processes user-provided strings (the video prompts) and interpolates them into bash commands (
infsh app run ...). While standard for this use case, it presents a theoretical surface for injection if theinfshCLI does not properly sanitize inputs.
Recommendations
- HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata