The Agent Skills Directory

[SAFE]: The skill serves as a structural and procedural guide for the HyperFrames Studio environment. It establishes conventions for timeline organization, such as ensuring each scene is a sub-composition and grouping elements by kind (video, graphics, captions, audio). These are legitimate development and design practices.
[EXTERNAL_DOWNLOADS]: The skill references an external repository at https://github.com/heygen-com/hyperframes-launches. This is an official GitHub repository for a well-known AI video platform, used here as a reference for project structure and workflow sequences. This reference is informational and does not trigger automatic execution of untrusted code.
[COMMAND_EXECUTION]: Instructions include running a local utility command hyperframes check to perform linting and verify project integrity. This is a standard quality assurance step for developers using this specific toolchain and does not involve arbitrary or dangerous command injection.
[INDIRECT_PROMPT_INJECTION]: The skill contains logic for processing user messages (e.g., distinguishing between a feature request and a general question). While it ingests user input to determine project changes, it defines high-level reasoning for the agent rather than exposing a vulnerability to malicious data processing. No unsafe interpolation of untrusted data into sensitive contexts was found.

hyperframes-studio