The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install additional functional skills from the belt-sh and inference-sh repositories and fetches a CLI installation guide from GitHub.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute belt CLI commands for account authentication, model searching, and triggering video generation processes.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user inputs, including text prompts and URLs for images, videos, and audio, which are then passed into the video generation pipeline.
Ingestion points: Text prompts and media asset URLs provided in the --input JSON parameter of the belt app run command.
Boundary markers: Command arguments are structured as JSON strings within the CLI call.
Capability inventory: The skill has the capability to execute shell commands via the Bash tool to interact with the external inference.sh API.
Sanitization: The skill does not perform local sanitization, relying instead on the safety filters and validation logic of the inference.sh platform.

seedance