The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for executing various media processing and project management tasks through the hyperframes CLI tool. All operations are scoped to the intended functionality of the tool, such as render, lint, validate, and init.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx to fetch and execute the hyperframes package from the npm registry. It also includes an upgrade command designed to check for and apply updates to the tool from its official source.
[INDIRECT_PROMPT_INJECTION]: The skill features capabilities to process external data, such as audio transcription (transcribe) and text-to-speech generation (tts). While these represent ingestion points for untrusted data, the operations are limited to media processing within the CLI environment and do not expose sensitive system APIs.
Ingestion points: Audio/video files via transcribe and text strings via tts in SKILL.md.
Boundary markers: None explicitly defined in the CLI command strings.
Capability inventory: Media rendering, file writing (via init and render), and environment checks (doctor).
Sanitization: Relies on the underlying hyperframes CLI and Whisper/TTS model implementations to handle input safely.

hyperframes-cli