llm-models
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the 'belt' CLI tool through the command 'npx skills add belt-sh/cli' and references documentation hosted on GitHub at 'inference-sh/skills'. These resources are associated with the primary functionality of the skill.
- [COMMAND_EXECUTION]: The skill is configured to use the 'belt' CLI tool via the Bash shell to perform tasks such as user authentication ('belt login') and running inference on specific LLM applications ('belt app run').
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided text prompts through JSON input which are then passed to external language models.
- Ingestion points: Data is ingested via the '--input' parameter in the 'belt app run' command within SKILL.md.
- Boundary markers: The skill uses structured JSON objects to separate the prompt from other command arguments.
- Capability inventory: The skill has the capability to execute the 'belt' CLI and interact with remote APIs for model inference.
- Sanitization: Content is passed directly to the inference tool, which typically relies on the downstream LLM provider's internal safety filters.
Audit Metadata