The Agent Skills Directory

[SAFE]: No malicious behavior, obfuscation, or safety bypasses were detected in the skill instructions.
[COMMAND_EXECUTION]: The skill uses the runcomfy CLI to process video generation requests, which is the stated primary purpose of the skill.
[EXTERNAL_DOWNLOADS]: Mentions the installation of the @runcomfy/cli package from npm and the skill itself from the agentspace-so/runcomfy-skills repository.
[SAFE]: The skill manages a surface for indirect prompt injection through user-supplied prompts and media URLs, implementing several security mitigations.
Ingestion points: User-provided prompt, image_url, video_url, and audio_url fields are processed by the skill in SKILL.md.
Boundary markers: User inputs are encapsulated within a JSON string for the CLI command.
Capability inventory: The skill invokes the runcomfy CLI via a subprocess call as described in SKILL.md.
Sanitization: The documentation explicitly states the CLI transmits the JSON body via HTTPS without shell expansion of the prompt content.

seedance-v2