The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requests access to the Bash tool but uses the allowed-tools configuration to restrict its execution capabilities solely to the belt command. This implementation of the principle of least privilege ensures that the agent cannot execute arbitrary shell commands outside the scope of the intended tool.
[EXTERNAL_DOWNLOADS]: The skill references external resources, including installation scripts and documentation, which point to the vendor's official GitHub repositories and domains. These references are consistent with the skill's functionality and trace back to the author's infrastructure.
[PROMPT_INJECTION]: Example commands demonstrate how to pass user prompts to LLMs via JSON inputs. The skill does not contain instructions that attempt to bypass safety filters or override agent constraints.

llm-models