youtube-thumbnail-design
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute belt CLI commands for image generation tasks via the inference.sh platform.
- [EXTERNAL_DOWNLOADS]: The documentation instructs the user to install the belt CLI and additional related skills from external GitHub repositories and npm packages associated with inference.sh.
- [PROMPT_INJECTION]: The skill features an attack surface for indirect prompt injection where user-supplied image descriptions are interpolated into shell commands executed via the belt CLI.
- Ingestion points: User-provided descriptions for thumbnail content are processed in the instructions within SKILL.md.
- Boundary markers: No explicit delimiters or boundary markers are used to isolate user-supplied prompt text from the rest of the shell command.
- Capability inventory: The skill possesses shell execution capabilities through the Bash tool used to run the belt CLI.
- Sanitization: No sanitization, escaping, or validation of user-provided prompt strings is specified before they are embedded into the JSON payload of the shell command.
Audit Metadata