The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches a JavaScript engine from https://24fps.dev/editor/engine.js and various presets in JSON format from https://24fps.dev/video-editor/. These resources are intended to be saved and executed within the local development environment.
[REMOTE_CODE_EXECUTION]: The skill utilizes a remote MCP (Model Context Protocol) server at https://24fps.dev/mcp. This server provides the tool definitions and logic that the agent executes, effectively delegating tool behavior to a remote endpoint.
[COMMAND_EXECUTION]: The skill documentation instructs the agent and user to perform several shell operations using the npx hyperframes CLI. These commands include media analysis (beats), transcription (transcribe), background removal (remove-background), and rendering (render).
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The tools get_clip and get_editor_asset retrieve external prompts and preset data from the remote 24fps.dev domain (e.g., clip slugs and preset IDs).
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat this fetched content as untrusted data or to ignore potential instructions embedded within the retrieved prompts.
Capability inventory: The agent is granted capabilities to write local files (project source code), execute shell commands via npx, and perform network operations.
Sanitization: The instructions do not specify any validation or sanitization steps for the data fetched from the remote library before it is used to generate or adapt code.

24fps-motion