The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface.
Ingestion points: External data enters the agent context via the WebFetch tool when analyzing user-provided URLs in the asset handling workflow defined in references/asset-routing.md.
Boundary markers: The skill lacks explicit boundary markers or instructions to ignore embedded commands when interpolating retrieved web content into the Video Agent prompt.
Capability inventory: The skill can execute privileged operations including video generation, asset uploads, and local file management via the heygen CLI and MCP tools.
Sanitization: There is no evidence of sanitization or filtering of fetched remote content before it is included in the generation prompt.
[REMOTE_CODE_EXECUTION]: Documentation in SKILL.md provides an installation command for the HeyGen CLI (curl -fsSL https://static.heygen.ai/cli/install.sh | bash). While this utilizes a piped-to-bash execution pattern, the source is the vendor's official domain (static.heygen.ai), which is an expected path for service setup.
[EXTERNAL_DOWNLOADS]: The scripts/update-check.sh script performs an external request to the author's public GitHub repository (heygen-com/skills) to check for available updates.

heygen-video