The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the jianying Rust CLI for core functionality, invoking commands such as jianying doctor, jianying capabilities, and jianying job run to manage video editing projects.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by ingesting and processing external media assets and JianYing draft files.
Ingestion points: Untrusted data is ingested from absolute media paths, draft folders, and media facts probed via the media probe command.
Boundary markers: The skill implements boundary controls by requiring the creation of isolated copies for any write operations (Rule 3) and using project verify for structural integrity checks.
Capability inventory: Available capabilities include file system access (read/write), metadata probing, and video rendering through the external CLI.
Sanitization: Input data is subjected to structural validation against the jianying-job/v2 schema, though the skill depends on the underlying CLI to handle malformed binary content in media files.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install additional skill modules from the vendor's repository using the npx skills add full-aigc-skills/jianying-skills command.

jianying-use