The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides links to installation instructions and setup scripts for the belt CLI, hosted on the official GitHub repository for inference.sh.
[COMMAND_EXECUTION]: The skill documentation guides the user to execute shell commands using the belt CLI tool for authentication and running model inference tasks.
[INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing user-defined prompts which are then sent to external LLMs. This represents a standard vector for indirect prompt injection.
Ingestion points: The --input flag in belt app run examples in SKILL.md allows passing arbitrary prompt text to the agent.
Boundary markers: Input data is encapsulated within JSON objects, which provides a structural boundary between instructions and data.
Capability inventory: The skill uses the Bash tool to execute belt commands.
Sanitization: No explicit input validation or prompt filtering is performed within the skill instructions.

llm-models