The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the @runcomfy/cli package from the npm registry. This is a standard installation of a vendor-provided tool.
[COMMAND_EXECUTION]: The skill uses the runcomfy CLI to interact with remote image models. It passes complex inputs, such as prompts and image URLs, as a JSON string via the --input flag, which reduces the risk of shell-based command injection.
[SAFE]: Security best practices are followed for credential management. The skill documentation advises storing API tokens in a configuration file with restricted permissions (0600) or using environment variables for CI/CD environments.
[SAFE]: The underlying CLI includes built-in safeguards, such as a 2 GiB file size cap for downloads, to prevent potential resource exhaustion issues.
[SAFE]: The skill defines a clear boundary for indirect prompt injection risks by processing third-party image URLs as untrusted data and utilizing structured JSON for prompt delivery.

relight