The Agent Skills Directory

[SAFE]: The skill functions as a documentation and instruction suite for the HeyGen Video Agent API, providing legitimate tools and examples for video generation.
[SAFE]: Authentication is handled securely through the use of environment variables (HEYGEN_API_KEY), avoiding hardcoded secrets and following industry standards for secret management.
[SAFE]: Network activity is confined to known, official HeyGen endpoints required for uploading assets, triggering generations, and downloading results.
[SAFE]: File system interactions are limited to the expected behaviors of reading media assets for upload and saving finished video files to local storage.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface where instructions embedded in user-provided text could influence the video generation process.
Ingestion points: The prompt parameter in mcp__heygen__generate_video_agent and direct API calls in video-agent.md.
Boundary markers: None present; the skill treats the user prompt as raw input for the Video Agent.
Capability inventory: The skill can read local files for assets (assets.md), perform network requests to the HeyGen API (video-agent.md), and write files to the local disk (video-status.md).
Sanitization: No explicit sanitization or escaping is performed on the user prompt before it is passed to the HeyGen API.

create-video