Deepgram Skills
Official agent skills for Deepgram speech-to-text, TTS, and voice agents.
About
Deepgram Skills is the official skills repository from Deepgram, covering its speech-to-text, text-to-speech, voice agent, and audio intelligence APIs. The 14 shipped skills include speech-to-text, text-to-speech, voice-agent, audio-intelligence, text-intelligence, api, cli, docs, examples, recipes, starters, browser-agent, setup-mcp, and self-hosted. Some are hand-written and others are generated from Deepgram's public OpenAPI and AsyncAPI specs, with a validation script that diffs the marketplace manifest against the filesystem.
The skills are agent-agnostic plain markdown consumable by Claude Code, Cursor, Windsurf, GitHub Copilot, Gemini CLI, and others, and the repo ships a Claude Code plugin-marketplace manifest. Deepgram maintains two STT families (Nova and Flux) and two TTS families (Aura and Flux) on separate endpoints, and the skills encode the decision guides for choosing between them. As with any generated API surface, verify endpoint behavior against the live docs when specs drift.
What it can do
- STT with Nova and Flux models
- TTS with Aura and Flux voices
- Voice agent and audio intelligence patterns
Setup
Install via the Claude Code plugin marketplace or copy the skill folders into your agent's skills directory.
Prerequisites
A Deepgram API key for live use.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 77; Trending score 80/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.