The Agent Skills Directory

[SAFE]: The skill restricts command execution to the runcomfy tool using platform configuration and provides clear guidance on secure token handling and installation.
[EXTERNAL_DOWNLOADS]: The skill references official vendor packages and service endpoints, including the @runcomfy/cli package and the runcomfy.com domain, which are required for the skill's primary purpose.
[PROMPT_INJECTION]: The skill documentation acknowledges the potential for indirect prompt injection from untrusted media URLs (e.g., character images or reference videos) and provides mitigation instructions, such as limiting ingestion to user-provided URLs. (Internal Severity: LOW)

controlnet-pose