The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to perform network requests to a user-provided API endpoint for key generation and retrieval.
[COMMAND_EXECUTION]: Utilizes a python3 one-liner to parse JSON responses from the proxy server and extract metadata.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data (such as key aliases, model names, and budget limits) which are then interpolated into a shell command payload.
Ingestion points: User-provided inputs for key configuration parameters (SKILL.md).
Boundary markers: None provided in the command templates.
Capability inventory: Network access via curl and local execution via python3 (SKILL.md).
Sanitization: No specific sanitization or escaping logic is defined within the skill instructions; it relies on the agent's default handling of variables.
[EXTERNAL_DOWNLOADS]: References official documentation at litellm.vercel.app, which belongs to a well-known hosting service and is associated with the project's vendor.

add-key