The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data via video_url and callback_url parameters. Additionally, the instruction field in the advanced API request provides a surface where malicious instructions could be embedded and passed to the HeyGen service.
Ingestion points: video_url, callback_url, instruction fields in SKILL.md examples.
Boundary markers: None present in the instructions to prevent the model from following instructions found in metadata or video content.
Capability inventory: The skill performs network operations via fetch and requests to api.heygen.com.
Sanitization: No evidence of sanitization or validation of the input URLs or instruction strings before they are sent to the external API.

video-translate