The Agent Skills Directory

[SAFE]: The skill serves as a legitimate workflow for video production, specifying clear steps for briefing, routing, and quality assurance.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the hyperframes component from heygen-com. This is a well-known service in the AI video industry, and the reference is documented as part of the skill's setup requirements.
[COMMAND_EXECUTION]: The instructions include usage of npx for skill updates and the Higgsfield CLI for image generation. These are standard tool invocations for the intended purpose of the skill and do not involve arbitrary or hidden shell execution.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface for indirect prompt injection through external inputs.
Ingestion points: The skill accepts external product descriptions and URLs during the brief stage in SKILL.md.
Boundary markers: There are no explicit delimiters defined for wrapping user-provided data.
Capability inventory: The skill has the capability to execute npx commands and interact with rendering engines.
Sanitization: The risk is mitigated by a mandatory verification step in the workflow where the agent must extract and review specific frames for visual audit before handoff.

ad-video