A dedicated tool for accessing a shared Whisper STT server over a private Tailnet using OpenAI-compatible endpoints.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install whisper-tailnet-api
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install whisper-tailnet-api using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Whisper Tailnet API?
The whisper-tailnet-api provides AI agents with a streamlined way to process audio files into text using a remote Whisper instance. By leveraging Openclaw Skills, developers can integrate high-performance speech-to-text capabilities directly into their agentic workflows without local resource overhead.
This skill is optimized for private network environments using Tailscale, ensuring secure and efficient communication between the agent and the transcription server while maintaining compatibility with standard OpenAI API patterns. It allows for quick health checks and high-fidelity transcription requests over a shared infrastructure.
Whisper Tailnet API Use Cases
- Remote transcription of audio logs or voice commands within a private network environment.
- Troubleshooting transcription accuracy or latency using standardized API calls.
- Validating speech-to-text response shapes for automated AI agent integration.
- Benchmarking transcription performance for different audio formats like WAV, MP3, or M4A.
How Whisper Tailnet API Works
- The AI agent performs a health check to ensure the Tailnet endpoint is reachable and responsive.
- The user or agent prepares an audio file and identifies its file extension.
- A POST request is sent to the transcription endpoint with the raw binary data as the body.
- The server processes the audio using the Whisper model (e.g., turbo).
- A JSON response containing the transcribed text and model metadata is returned to the agent for downstream processing.
Whisper Tailnet API Setup
Ensure you are connected to the relevant Tailnet and the Whisper server is online at the designated IP. You can verify the connection using the health check endpoint.
# Verify connectivity
curl -sS http://100.92.116.99:8765/health
To transcribe an audio file, use the following command structure within your Openclaw Skills workflow:
curl -sS -X POST \
--data-binary @/path/to/audio.wav \
"http://100.92.116.99:8765/v1/audio/transcriptions?ext=.wav"
Whisper Tailnet API Data Schema & Taxonomy
The API interacts with binary audio data and returns structured JSON metadata. This helps agents parse results consistently.
| Parameter | Type | Location | Description |
|---|---|---|---|
ext |
string | Query | File extension (e.g., .wav, .mp3, .m4a) |
language |
string | Query | Optional: Language hint to improve accuracy |
Response Format
| Field | Type | Description |
|---|---|---|
text |
string | The generated transcription string |
model |
string | The specific Whisper model version used for the task |
Whisper Tailnet API Advanced Features
- OpenAI-compatible routing allows for drop-in replacement in existing AI pipelines and Openclaw Skills libraries.
- Language hint support to significantly improve transcription accuracy for multi-lingual audio inputs.
- Support for timing tests to monitor network and inference latency in real-time.
- Direct binary body processing which reduces payload overhead compared to traditional multipart/form-data uploads.
SKILL.md
Loading
Related Openclaw Skills
Local OpenAI-Compatible TTS
Enable high-quality, local text-to-speech with cloned voices in your OpenClaw instance using OpenAI-compatible backends.
Cloudflare Whisper Worker
A high-performance audio transcription skill leveraging Cloudflare Workers and the Whisper model for seamless speech-to-text conversion.
ShellPhone Gateway
A private WebSocket gateway that establishes a secure, self-hosted bridge between iOS devices and local AI agents.
Reformed Books Search
A specialized AI agent skill for searching and downloading Reformed Christian theology books and academic PDF resources.
ADHD Assistant
A comprehensive executive function support system designed to help users manage daily planning, task breakdown, and emotional regulation through evidence-based ADHD strategies.
ADHD Daily Planner
A neurodivergent-friendly AI skill that provides flexible planning, realistic time estimation, and executive function support for ADHD brains.
METADATA
Featured*
