juce-agent-toolkit

Create and inspect audio projects in the JUCE Agent Toolkit Harness workspace.

465|41|Updated Aug 4, 2026
One-click install
npx skills add https://github.com/autonomous-ai/openharness --skill juce-agent-toolkit-autonomous-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: juce-agent-toolkit
Source: https://github.com/autonomous-ai/openharness/tree/main/store/agents/juce-agent-toolkit/skills/juce-agent-toolkit
Command: npx skills add https://github.com/autonomous-ai/openharness --skill juce-agent-toolkit-autonomous-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Building and validating audio instruments requires iterating on DSP parameters, rendering test audio, and comparing results, which is tedious to do manually across preview renderers and native JUCE builds. ## Core Features & Use Cases - Instrument creation and validation: Modify audible properties like waveform, pitch, attack, or brightness, render a four-note phrase before and after, and compare peak/RMS levels between WAV files. - Native JUCE rendering: Edit the workspace CMake project's Source/main.cpp, run the juce action to compile a native offline renderer, and inspect the measured recording. - Run tracking and reproducibility: Successful artifacts and measurements are stored in out/runs/<id>/ with out/latest.json naming the current result, while failed runs preserve the last success and record the error. - Use Case: An instrument maker adjusts the brightness control in studio.json, renders the same phrase before and after the change, and tells the user which run contains the result and what was actually measured. ## Quick Start Ask the agent to change one audible property such as waveform or brightness in the JUCE workspace, render the four-note phrase before and after, and compare the resulting WAV files.

Frequently Asked Questions about juce-agent-toolkit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create and validate an instrument with the JUCE Agent Toolkit?

Change one audible property such as waveform, pitch, attack, or brightness in studio.json, then render the same four-note phrase before and after. Compare peak/RMS levels and listen to both WAV files to validate the change.

How do I render audio with the JUCE workspace toolchain?

Run "$STUDIO_TOOLCHAIN/run.sh" render to produce a new result. Successful artifacts and measurements land in out/runs/<id>/, and out/latest.json names the current result.

Can the JUCE starter renderer export a VST3 plugin?

No, the starter renderer is not a VST3 plugin. For full DAW plugins, read the juce-project-create, juce-project-starter, and juce-build-release skills under $STUDIO_UPSTREAM, which require their own host and audio checks.

What happens when a JUCE render run fails?

A failed run preserves the last successful result and records the error in the verdict. The previous success remains available in out/runs so no working artifact is lost.

What is the difference between the browser audition and the native JUCE renderer?

The browser audition and Python WAV renderer are local DSP previews, while the JUCE action compiles a native offline renderer from the workspace CMake project. Preview results should be distinguished from native service or hardware output.