The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run localized development and build scripts (e.g., bun generate.ts, bun run build, and bun packages/docs/generate-sfx-waveforms.ts). It also references using system utilities like afinfo and ffmpeg for media processing. These operations are constrained to the project environment and serve the skill's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided metadata for sound effects to generate documentation and code exports.
Ingestion points: External metadata such as fileName and attribution are used to populate generate.ts entries and MDX documentation files.
Boundary markers: The instructions do not define explicit delimiters for untrusted input, relying instead on the agent following existing file patterns.
Capability inventory: The skill involves writing to multiple project files (TypeScript, MDX, and Markdown) and executing media processing scripts via bun and ffmpeg.
Sanitization: There are no explicit instructions for sanitizing or escaping user-supplied strings before they are interpolated into the code and documentation.
[DATA_EXPOSURE]: The skill refers to the .env file within the repository as a prerequisite for deployment. This is a standard operational note regarding project configuration and does not include patterns for reading or exfiltrating credentials.

add-sfx