The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs file system operations by creating directories and downloading video content to a user-specified path.
Evidence: In scripts/video_generate.py, the urllib.request.urlretrieve function is used to save the generated video to a filename provided via command-line arguments.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and image data which is then passed to an external model API, creating a surface for potential indirect instructions.
Ingestion points: The prompt and first_frame arguments in scripts/video_generate.py accept arbitrary strings and file paths.
Boundary markers: Absent; inputs are not wrapped in delimiters or accompanied by instructions to ignore embedded commands.
Capability inventory: The skill possesses network access (Ark API and urllib), file read capabilities (for image encoding), and file write capabilities (urlretrieve).
Sanitization: Absent; input data is transmitted directly to the model endpoint without filtering.
[DATA_EXPOSURE]: The skill reads authentication credentials from environment variables to interact with the Volcano Engine Ark service.
Evidence: scripts/video_generate.py accesses MODEL_VIDEO_API_KEY, ARK_API_KEY, VOLCENGINE_ACCESS_KEY, and VOLCENGINE_SECRET_KEY for API authentication.

video-generate