The Agent Skills Directory

[SAFE]: The skill uses a Python script (scripts/videogen.py) to interact with the AI Hive OpenAPI. All network requests are directed to the service domain ai-hive.iclip.cn or authorized object storage URLs for media handling.
[DATA_EXFILTRATION]: The skill manages an API key stored in ~/.ai-hive/config.json. The script correctly implements security best practices by setting the file permissions to 0600 (read/write by owner only) to protect the stored credentials from unauthorized access by other users on the system.
[PROMPT_INJECTION]: The skill processes user-supplied creative prompts to generate videos through a remote API. Ingestion points: User input provided via the --prompt argument in videogen.py. Boundary markers: The prompt is encapsulated within a JSON object for API transmission. Capability inventory: The script is limited to API interaction and file downloads; it does not execute code based on model responses. Sanitization: Standard JSON serialization is used to prevent injection into the request structure.
[EXTERNAL_DOWNLOADS]: The skill requires the requests library, which is a standard and well-known Python package for HTTP operations. It does not download or execute arbitrary scripts from untrusted sources.

dynamic-comic-video