Menu
Audio transcription (Whisper) and text-to-speech (OpenAI TTS). Use for processing audio messages or generating voice responses.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Clappie is a digital assistant layer that turns Claude Code into a full personal assistant with interactive terminal UIs called "clapps". It manages everything through one CLI: `clappie`. This is the core skill for the entire project. Load it whenever the user asks for ANYTHING personal-assistant related โ not just when they say "clappie". Examples: emails, texts, notifications, inbox, dashboard, sidekicks, chores, heartbeat, background apps, displays, parties, memory, messages, todos, approval queue, status, automation, or any `[clappie]` prefixed message. If the user says "open notifications" or "spawn a sidekick" โ that's this skill. Don't guess at how these systems work โ load this skill, it has the docs. CORE SYSTEM - Terminal display engine with push/pop view navigation, mouse click handling, keyboard shortcuts, and two-way communication. Views send structured messages back to Claude prefixed with [clappie]. SUBSYSTEMS (each has deep docs in REFERENCE.md files): - Sidekicks: Autonomous AI agents runni
Slack bot integration for receiving messages and reactions. Use when setting up Slack webhooks or managing Slack bot configuration.
Telegram bot integration for receiving messages and reactions. Use when setting up Telegram webhooks or managing Telegram bot configuration.
| name | whisper |
| description | Audio transcription (Whisper) and text-to-speech (OpenAI TTS). Use for processing audio messages or generating voice responses. |
Listen to audio, say things out loud. Powered by OpenAI's Whisper and TTS.
Requires OPENAI_API_KEY in your .env file.
# Listen to audio (transcribe)
clappie whisper listen /path/to/audio.ogg
# Say something (text-to-speech)
clappie whisper say "Hello world"
# Say with specific output path
clappie whisper say "Hello world" -o recall/files/hello.opus
# Pick voice by mood
clappie whisper say "I'm feeling chaotic" -m chaotic
# Use specific voice
clappie whisper say "Serious business" -v onyx
# List all voices
clappie whisper voices
Audio files go to recall/files/ with everything else. Override with -o if needed.
| Voice | Vibe |
|---|---|
| alloy | Neutral, balanced |
| echo | Male, warm |
| fable | British male, expressive |
| onyx | Deep male, authoritative |
| nova | Female, friendly |
| shimmer | Soft female, gentle |
When you use -m mood, a matching voice is picked:
When receiving voice messages in a sidekick:
# Listen to what they said
clappie whisper listen recall/files/voice-xxx.ogg
# Say something back
clappie whisper say "Your response here" -m playful
# Output lands in recall/files/say-<timestamp>.opus
clappie sidekick send-file voice recall/files/say-<timestamp>.opus
-s (0.25 to 4.0)