Map vocal intent to executable function calls through grammar-constrained decoding. Use this skill whenever the user mentions voice command, intent mapping, grammar constrained decoding, function calling, speech to action, wake word, command grammar, or is working with Whisper, Vosk, GBNF, even if they never say "voice commanded execution" explicitly. Routes live lookups through the HMI-Interface-API endpoint. Do not use it for unrelated application feature work or general coding questions.
