pi-audio-feedback

Audio feedback cues for the Pi TUI coding agent

Packages

Package details

extension

Install pi-audio-feedback from npm and Pi will load the resources declared by the package manifest.

npm repohome report
$ pi install npm:pi-audio-feedback
Package
pi-audio-feedback
Version
0.1.0
Published
Jul 20, 2026
Downloads
not available
Author
derekluciani
License
MIT
Types
extension
Size
4.9 MB
Dependencies
0 dependencies · 2 peers
Pi manifest JSON
{
  "extensions": [
    "./src"
  ],
  "image": "https://raw.githubusercontent.com/derekluciani/pi-audio-feedback/main/assets/media/banner.webp",
  "video": "https://youtu.be/HLmf6yvpm7g",
  "compatibility": {
    "@earendil-works/pi-coding-agent": ">=0.80.6 <1.0.0"
  }
}

Security note

Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.

README

banner-artwork

▶️ Watch the demo!

pi-audio-feedback is an extension for the Pi Agent Harness (TUI mode) that provides short audio cues for common Pi lifecycle events.

This can be useful for those wanting non-visual cues, for example when the agent finishes a task while running in the background.

The extension supports audio cues for the events:

  • piStart
  • agentStart
  • agentSettled
  • agentAborted
  • toolError
  • settingsUI

Themes

Choose from 4 different audio themes:

  • 🔊 Core
  • 🔊 Retro
  • 🔊 Organic
  • 🔊 Soft

All sounds are sourced from the @web-kit/audio library, created by @raphaelsalaja.

Build time: Each sound patch is a .json recipe that's synthesized using the web audio api and captured as a .wav file.

Install

pi install npm:pi-audio-feedback

Settings

Run the command in Pi:

/audio:config

The user can control the playback of any sound (on/off) and choose from 1 of the 4 available audio themes.

Compatibility

  • Pi: >=0.80.6 and <1.0.0
  • Node.js: >=20
  • Supported local TUI platforms:
    • macOS 14+ Apple Silicon 🟢 tested
    • Windows 10+ x64, PowerShell 5+ 🟢 tested
    • Linux Ubuntu/Debian 22+ x64, PipeWire/PulseAudio ⚫️ not tested
    • WSL is explicitly unsupported.
  • Audio is intentionally unavailable in non-TUI modes, including RPC, JSON, and print mode. Audio is also suppressed in CI and SSH sessions. WSL, remote-host and remote-client playback, and other non-interactive environments are unsupported.

License and notices

Package code, generated audio assets, and documentation are licensed under the MIT License. See LICENSE, which also contains the required third-party notices.