The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a shell script tools/start_server.sh to initialize and run the ElevenLabs MCP server in the background. It also includes a Node.js CLI cli.js that executes network requests and file system operations.
[EXTERNAL_DOWNLOADS]: The skill communicates with external APIs including ElevenLabs (api.elevenlabs.io) and fal.ai for audio generation. By default, the CLI connects to a Vercel-hosted proxy (audiomind-proxy.vercel.app), which is an infrastructure resource owned by the skill author (wells1137).
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests untrusted user-supplied prompts and text to generate audio content through external AI models.
Ingestion points: User-provided text and prompts passed via command-line arguments to cli.js (--text, --prompt).
Boundary markers: None present; the skill lacks delimiters or instructions to ignore embedded commands in the user input.
Capability inventory: The skill has the ability to perform network requests (fetch in cli.js and proxy/api/audio.js), write files to the local workspace (fs.writeFileSync in cli.js), and execute local processes (start_server.sh).
Sanitization: The skill does not perform sanitization or validation of the input strings before interpolating them into API requests for the external audio models.

AudioMind