The Agent Skills Directory

[SAFE]: The skill uses the allowed-tools: Bash(belt *) configuration to enforce the principle of least privilege, ensuring that the AI agent can only execute the specific belt command and cannot run arbitrary shell commands.
[EXTERNAL_DOWNLOADS]: The skill references documentation and installation scripts from raw.githubusercontent.com/inference-sh/, which corresponds to the official GitHub organization for the documented inference.sh service. These references are documented neutrally as they originate from the tool's official source.
[INDIRECT_PROMPT_INJECTION]: The skill provides templates for interpolating user-provided text into command arguments for image generation. This represents an ingestion surface for untrusted data, but the risks are mitigated by the explicit tool-scoping restrictions and the structured JSON format used for inputs.

product-photography