The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes workflows for reverse-engineering editing styles by analyzing external video content and social media posts. While this creates a potential surface for indirect prompt injection, the risk is mitigated by the structural nature of the task.
Ingestion points: Video transcripts and social media metadata ingested via the watch-video and social-fetch tools as described in the edit-anatomy reference.
Boundary markers: Absent; the skill does not explicitly instruct the agent to ignore instructions embedded within the ingested transcripts.
Capability inventory: Access to shell commands (npm/npx) and external API interactions via the HeyGen MCP server.
Sanitization: Absent; no explicit sanitization or filtering of ingested text is mentioned.
[EXTERNAL_DOWNLOADS]: References the installation of established programmatic video libraries like hyperframes and Remotion. It also links to official documentation for reputable AI services including HeyGen, Runway, and Google's Veo. These resources are from trusted organizations and well-known services.
[COMMAND_EXECUTION]: Provides standard setup instructions for developers, such as running npm install and npx create-video. These are necessary for the skill's primary purpose and follow standard industry practices for initializing software projects.

video