Skip to content

SKILL.md

Latest commit

 

History

History
107 lines (82 loc) · 2.76 KB

SKILL.md

File metadata and controls

107 lines (82 loc) · 2.76 KB
Raw
name faye-voice
description Install and operate Faye voice for OpenClaw: always-on wake-word listener, ElevenLabs speech output, local dashboard/API profile manager, and service controls for macOS/Linux. Use when setting up voice sessions, managing wake-word profiles, hardening listener reliability, or troubleshooting voice flow between terminal and Telegram/OpenClaw.

Faye Voice Skill

Faye gives an OpenClaw agent an always-on voice interface with custom wake words and profile switching.

Capabilities

  1. Always-on wake-word listener service
  2. ElevenLabs TTS playback via local or remote speaker scripts
  3. Local dashboard for profile management and one-click activation
  4. Local API + SSE event stream for live status and wake/session events
  5. Optional Telegram bridge service for OpenClaw command return path (#faye_speak)

Setup

1. Install

./scripts/install.sh

2. Configure / Update

./scripts/faye setup

3. Open Dashboard

open http://127.0.0.1:4587

Operations

Profile Management

./scripts/faye profile list
./scripts/faye profile activate --id <profile-id>
./scripts/faye profile create --name "Desk" --voice-id "..." --voice-name "..." --wake-word "Faye Arise"

Voice Playback

./scripts/speak.sh "Testing voice output"
./scripts/speak-remote.sh "Remote playback test"

Health and Service Control

./scripts/faye doctor
./scripts/listener-control.sh status
./scripts/dashboard-control.sh status
./scripts/telegram-bridge-control.sh status

Config Contracts

  • Runtime config: ~/.openclaw/faye-runtime-config.json
  • Active legacy config: ~/.openclaw/faye-voice-config.json
  • ElevenLabs key: ~/.openclaw/secrets/elevenlabs-api-key.txt
  • Local event token: ~/.openclaw/secrets/faye-local-event-token.txt

API Contracts (Local only)

  • GET /v1/health
  • GET /v1/profiles
  • POST /v1/profiles
  • PATCH /v1/profiles/:id
  • DELETE /v1/profiles/:id
  • POST /v1/profiles/:id/activate
  • POST /v1/speak
  • POST /v1/speak/test
  • GET /v1/events
  • POST /v1/listener/restart
  • POST /v1/bridge/restart

Files

  • scripts/install.sh

  • scripts/faye

  • scripts/install-listener.sh

  • scripts/install-dashboard.sh

  • scripts/install-telegram-bridge.sh

  • scripts/telegram-bridge-control.sh

  • scripts/listener.sh

  • scripts/speak.sh

  • scripts/speak-remote.sh

  • scripts/install-speaker.sh

  • scripts/voice-picker.sh

  • scripts/seven-shadow-test.sh

  • .github/workflows/ci-quality.yml

  • .github/workflows/seven-shadow-system.yml

  • references/supported-voices.md

  • references/troubleshooting.md

  • references/openclaw-telegram-protocol.md

  • references/reliability-slo.md

  • references/seven-shadow-system.md

  • governance/seven-shadow-system