Skip to main content
SLNG skills are a pack of skills for your coding agent. Each skill is a folder your coding agent loads on demand, teaching it to use the SLNG API: speech to text, text to speech, voice agents, prompt authoring, and migrations for LiveKit, Pipecat, and custom stacks.

​
Prerequisites

  • Node 18 or later, for npx skills.
  • A coding agent that loads skills from a directory (Claude Code, Codex, Gemini CLI, and others).
  • An SLNG API key. See Create your API key.

​
Install

slng-ai/skills
Loading repository data...
 Install the pack with the skills CLI. It detects the coding agents you have installed, or prompts you to pick:
Skills install into your coding agent’s skills directory, for example ~/.claude/skills/. Target a specific coding agent with -a (claude-code, codex, or gemini-cli), and add -g to install for every project. Re-run the command any time to update. To install a single skill, name it:

​
Set up your API key

The skills read your SLNG API key from the environment. The core skills use VOICEAI_API_KEY; the migration skills use SLNG_API_KEY. Set the one the skills you install need, or set both to the same key.
The setup-api-key skill configures and validates this for you. See Create your API key to get a key.

​
Available skills

For the full list of models, see the model catalog: text to speech and speech to text. Your coding agent can also list the currently deployed models with voiceai models --tts or voiceai models --stt. Each skill shows the same workflow in four flavors: the voiceai CLI, the Python SDK, the TypeScript SDK, and direct REST or WebSocket calls.

​
Use a skill

Once a skill is installed, describe the task to your coding agent in plain language. The skill’s description tells it when to load the skill, so you do not call the skill by name.

​
Set up your key

The setup-api-key skill checks for your key and helps you set it.

​
Generate speech

The text-to-speech skill turns text into audio with a model and voice you name, or sensible defaults.

​
Transcribe audio

The speech-to-text skill transcribes a file or a live stream.

​
Draft a voice agent prompt

The agent-prompt skill drafts a system prompt, greeting, and tools for the SLNG Agent Builder.

​
Create and run a voice agent

The agents skill creates the voice agent and can start a web session or place an outbound call.

​
Migrate a project

The migration skills swap your speech to text and text to speech for SLNG, and can route the language model through the Context Router.

​
Next steps