The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data from a video timeline and clip visual context, creating an ingestion surface for potential indirect prompt injection attacks.

Ingestion points: Data is sourced from timeline clips, visual context, and the output of the analyze_shots tool.
Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded within the video project's metadata or filenames.
Capability inventory: The skill has the ability to modify the project state through tools such as removeEffect, updateEffect, setColor, and applyFilterPreset.
Sanitization: There is no evidence of sanitization, escaping, or filtering applied to external content before the agent processes it.

colourist