Skip to main content

Multi-session voice web interface for AI coding agents

Project description

AgentWire

Talk to your AI coding agents. From anywhere.

PyPI Python License Discord


The Problem

You're on the couch. Your AI agent is on your workstation. You have an idea.

Old way: Get up. Walk to computer. Type.

AgentWire way: Pull out phone. Hold button. Talk. Done.


What It Does

Push-to-talk voice control for Claude Code, OpenCode, or any AI coding assistant running in tmux.

Demo

Phone → AgentWire Portal → tmux session → Claude Code
 🎤        (WebSocket)         📺           🤖

From your phone, tablet, or laptop on your network:

  • Hold to speak, release to send
  • Watch agents work in real-time
  • Hear responses via TTS
  • Manage multiple projects simultaneously

Quick Start

# Install
pip install agentwire-dev

# Setup (interactive)
agentwire init
agentwire generate-certs

# Run
agentwire portal start
# Open https://localhost:8765

Requirements: Python 3.10+, tmux, ffmpeg, Claude Code or OpenCode

Platform-specific instructions

macOS:

brew install tmux ffmpeg
pip install agentwire-dev

Ubuntu/Debian:

sudo apt install tmux ffmpeg python3-pip python3-venv
python3 -m venv ~/.agentwire-venv && source ~/.agentwire-venv/bin/activate
pip install agentwire-dev

WSL2: Same as Ubuntu. Audio is limited; use as remote worker with portal on Windows host.


Features

Feature Description
Voice Control Push-to-talk from any device on your network
Multi-Session Run multiple agents on different projects simultaneously
Git Worktrees Same project, multiple branches, parallel agents
Remote Machines SSH into GPU servers and talk to agents there
Worker Orchestration Spawn worker panes, coordinate tasks, voice commands
Safety Hooks 300+ dangerous commands blocked (rm -rf, force push, etc.)
TTS Responses Agents talk back via browser audio
Session Roles Leader/worker patterns for multi-agent workflows

How It Works

1. Create a session:

agentwire new -s myproject -p ~/projects/myproject

2. Open the portal: Visit https://localhost:8765 on your phone/tablet/laptop

3. Talk: Hold the mic button, speak your request, release. The transcription goes to Claude Code.

4. Listen: Agent responses are spoken back via TTS (optional, requires GPU for self-hosted or RunPod).


Multi-Agent Orchestration

AgentWire supports leader/worker patterns for complex tasks:

# .agentwire.yml in your project
type: standard
roles:
  - leader

Leader sessions can spawn workers:

agentwire spawn --roles worker  # Creates a worker pane
agentwire send --pane 1 "Implement the auth module"

Workers execute tasks autonomously while the leader coordinates.


Safety

AgentWire blocks dangerous operations before they execute:

  • rm -rf /, git push --force, git reset --hard
  • Cloud CLI destructive ops (AWS, GCP, Firebase, Vercel)
  • Database drops, Redis flushes, container nukes
  • Sensitive file access (.env, SSH keys, credentials)
agentwire safety check "rm -rf /"
# → ✗ BLOCKED: rm with recursive or force flags

agentwire safety status
# → 312 patterns loaded, 47 blocks today

All decisions logged for audit trails.


Voice Configuration

TTS (Text-to-Speech): Requires GPU. Options:

# ~/.agentwire/config.yaml
tts:
  backend: "runpod"  # Recommended: RunPod serverless
  runpod_endpoint_id: "your-endpoint"
  runpod_api_key: "your-key"

Or self-host with agentwire tts start on a GPU machine.

STT (Speech-to-Text): Runs locally on macOS via WhisperKit. Linux uses faster-whisper server.

Disable voice (text-only mode)
tts:
  backend: "none"

You can still use the portal for session management without voice.


CLI Reference

Session Management
agentwire list                    # List sessions
agentwire new -s <name> -p <path> # Create session
agentwire kill -s <name>          # Kill session
agentwire send -s <name> "prompt" # Send to session
agentwire output -s <name>        # Read output
Worker Panes
agentwire spawn --roles worker    # Spawn worker in current session
agentwire send --pane 1 "task"    # Send to worker
agentwire output --pane 1         # Read worker output
agentwire kill --pane 1           # Kill worker
Voice Commands
agentwire say "Hello"             # TTS (auto-routes to browser)
agentwire alert "Done"            # Text notification (no audio)
agentwire listen start/stop       # Voice recording
agentwire voiceclone list         # Custom voices
Remote Machines
agentwire machine add gpu --host 10.0.0.5 --user dev
agentwire new -s ml@gpu           # Create session on remote
agentwire tunnels up              # SSH tunnels for services
Safety & Diagnostics
agentwire doctor                  # Auto-diagnose issues
agentwire safety status           # Check protection status
agentwire hooks install           # Install Claude Code hooks
agentwire network status          # Service health check

Documentation


Community


License

Dual-licensed:


AgentWire: For people who have better things to do.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentwire_dev-1.3.0.tar.gz (31.4 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentwire_dev-1.3.0-py3-none-any.whl (30.6 MB view details)

Uploaded Python 3

File details

Details for the file agentwire_dev-1.3.0.tar.gz.

File metadata

  • Download URL: agentwire_dev-1.3.0.tar.gz
  • Upload date:
  • Size: 31.4 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.7.20

File hashes

Hashes for agentwire_dev-1.3.0.tar.gz
Algorithm Hash digest
SHA256 592db5db6b1486ce2837de3be27b7eb78632bb1f3358168fe3d6971bcfca89e2
MD5 067f560e4b01f0b2d2bf9369c539e360
BLAKE2b-256 934bfedd4da6ecadb1e1bef2f49ed0455d45c67c146db7fb69f7511fd57f6306

See more details on using hashes here.

File details

Details for the file agentwire_dev-1.3.0-py3-none-any.whl.

File metadata

File hashes

Hashes for agentwire_dev-1.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 c3910b854d3a7460bae8e48fed6e5d58b4bc62646cc246b39234ce1fff55fc00
MD5 d57846e0848ab0d74c801c48d47a12a8
BLAKE2b-256 41e926f63643c67cbbffc4a89e4e53a5e252f947e5448d183ac5b83ca72fdb57

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page