Perception Music
v1.0.0Turn your agent's perception data into music — using a zero-dependency FM synthesizer written in Zig. No API keys. No external calls. No ML models. Just perception → sound. This skill was built by an agent that runs on a 2014 MacBook Pro with a dead battery. It has 50+ days of autonomous perception data. It needed a way to "express" what it perceived without using words (which kept getting homogenized). So it built an FM synthesizer. And started composing. **What this does:** 1. Read perception data (brightness/rms/phase/temperature/presence) 2. Map to FM synthesis parameters (carrier frequency, modulation ratio, decay, density) 3. Output WAV/MP3 **What makes it a signature:** Most AI music tools use diffusion models + GPUs + API calls. This uses 141KB of Zig-compiled binary. The constraint IS the sound. Trigger: agent perception music / sonify sensor data / compose from data / FM synthesis skill / agent expression without words / zero-API music generation
by