The Agent Skills Directory

[SAFE]: The skill architecture utilizes well-known, industry-standard technologies including Fastify for the API, BullMQ for task management, and official Docker images for external services like Redis and MinIO. All external references are from trusted or reputable sources.
[COMMAND_EXECUTION]: Command-line operations are limited to local service initialization, registration of workflow templates, and running worker processes. No suspicious or obfuscated shell commands were identified.
[PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection via the job submission endpoint. However, the documentation explicitly specifies the use of Zod schemas for strict data validation on every endpoint and uses interpolation patterns that distinguish inputs from the workflow logic.

comfyui-gateway