The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute curl commands. These commands are designed to interact with the LiteLLM proxy API to add new models and perform test completions.
[DATA_EXFILTRATION]: The skill prompts for and transmits sensitive information, including the LiteLLM proxy admin key and provider-specific API keys, to a user-provided base URL. This data transfer is the intended administrative function of the skill for configuring the LiteLLM proxy and is consistent with the author's infrastructure.
[PROMPT_INJECTION]: The skill instructions involve interpolating user-provided inputs into shell command templates, which represents a potential indirect injection surface.
Ingestion points: User input for model names, providers, and credentials specified in SKILL.md.
Boundary markers: None identified; the instructions provide a direct template for interpolation.
Capability inventory: Execution of network requests via curl through the Bash tool.
Sanitization: No specific sanitization or escaping instructions are provided for the user-supplied parameters.

add-model