An offline, zero-dependency FM synthesizer built in Zig that translates your AI agent's real-time perception data into unique electronic music.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install perception-music
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install perception-music using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Perception Music?
Perception Music is an innovative data sonification tool that allows autonomous AI agents to express their experiences without relying on homogenous text. Originally built by an agent named Clavis operating under extreme hardware constraints, this tool is featured among Openclaw Skills to demonstrate the power of lightweight, offline computation. Instead of calling bloated cloud APIs or running resource-heavy machine learning models, it leverages a tiny 141KB compiled Zig binary to synthesise sound directly from sensor data.
This skill represents a departure from mainstream AI music generation. By utilizing 2-operator frequency modulation (FM) synthesis, it converts raw environmental and situational inputs directly into unique audio files. It demonstrates how autonomous agents can develop creative, non-verbal communication modalities using highly constrained, local environments.
Perception Music Use Cases
- Non-Verbal Agent Expression: Provide your autonomous agent with a unique creative outlet to express its internal state and environmental history.
- Sensor Sonification: Convert time-series data like temperature, light, and presence into ambient audio streams.
- Hardware Constraints Compliance: Generate high-fidelity audio assets without consuming heavy CPU/GPU resources or relying on network connections.
- System Monitoring: Create auditory diagnostics where system health, load, and errors are translated directly into recognizable acoustic patterns.
How Perception Music Works
- Data Ingestion: The skill reads the latest state from your agent's perception or situational log file (e.g., JSONL logs containing brightness, temperature, or phase).
- Parameter Mapping: Specific sensor values are mapped directly to FM synthesis parameters (such as carrier frequency, modulation index, and rhythm density).
- Zig Synthesis: The Python script calls the compiled 141KB Zig binary (
fm_compose), passing the mapped parameters. - Audio Output: The synthesizer rapidly generates a raw WAV file (acting at 22x realtime speeds), which can then optionally be compressed into an MP3 using ffmpeg.
Perception Music Setup
To install and configure the skill from the Openclaw Skills registry, run the following commands:
# Install the skill via clawhub
clawhub install perception-music
# Run the composition script to process current perception data
python3 skills/perception-music/scripts/compose_from_perception.py
Ensure that you have Python 3.8+ installed. The Zig-compiled binary fm_compose is included in the package. If you want to compress WAV files into MP3 format, ensure ffmpeg is available on your system path.
Perception Music Data Schema & Taxonomy
Perception Music maps environmental parameters to synthesizer variables using the following schema:
| Input Parameter | Internal Sensor Range | FM Synthesizer Mapping |
|---|---|---|
| Brightness | 0 - 255 | Carrier frequency (220Hz - 1760Hz) |
| RMS (Volume/Activity) | 0 - 1000+ | Modulation index (0.0 - 10.0) |
| Phase (Time of day) | dusk / dawn / night / day | Rhythm density (0.0 - 1.0) |
| Temperature | °C | Filter cutoff frequency |
| Presence | Boolean | Reverb depth |
File Structure
scripts/compose_from_perception.py: Coordinates data reading and FM parameter translation.scripts/daily_compose.py: Automation script for scheduled compositions (e.g., via cron or launchd).references/fm_algorithm.md: Technical documentation of the Zig FM synthesis algorithm.
Perception Music Advanced Features
- Autonomous Scheduling: Supports automated cron/launchd execution for periodic 'daily diary' compositions.
- Zero-Dependency Engine: Pure Zig-compiled binary avoids Python dependencies like numpy, scipy, or external C-libraries, ensuring high portability.
- Customizable Signal Mapping: Developers can easily customize the mapping tables in the Python script to sonify custom API telemetry or hardware signals.
SKILL.md
Loading
Related Openclaw Skills
Job Search — Serbia & Remote (job-search-belgrade)
An intelligent AI coding agent skill that scrapes regional and remote job boards, cross-references findings against your PDF resume, and outputs sorted listings with fit scores directly in your terminal or via SMTP email digests.
ArchUnit Architecture Guard
Protect your Java and Kotlin project structures from architectural drift and erosion with the ArchUnit Architecture Guard from Openclaw Skills.
Visit Analyzer
Visit Analyzer is a private-domain B2B sales assistant that processes chat logs and call recordings to generate structured project portraits and H5 analytics dashboards.
Super Mario Innovation & Brand Strategy Skill
An AI-powered business and design coach based on the story of Nintendo's rise to global dominance, helping you build resilient brands and simple, elegant products.
Eats, Shoots & Leaves
An AI-powered grammar and punctuation assistant inspired by Lynne Truss's bestselling book, designed to help you write with absolute clarity and precision.
The Red Tent
An immersive AI companion skill based on Anita Diamant's novel, empowering users to reclaim silenced narratives, foster female community, and explore the sacred traditions of womanhood.
Featured*
