Providers
Zeno uses an AI provider to analyze your code. On first run, you will be prompted to choose a provider and enter an API key. Your choice is saved to ~/.zenoai/config.json.
Supported providers
| Provider | Where to get a key |
|---|---|
| Anthropic | console.anthropic.com |
| OpenAI | platform.openai.com |
| Gemini | aistudio.google.com |
| OpenRouter | openrouter.ai |
Switching providers
To switch providers or update your API key, run:
npx zenoai resetThis clears your saved config. The next npx zenoai run will prompt you to choose a provider and enter a new key.
Provider notes
Anthropic — Zeno was developed and tested most extensively against Anthropic models. Recommended if you do not have a preference.
OpenAI — Supported.
Gemini — Supported via the Gemini API.
OpenRouter — Supported for routing to multiple model providers through a single API key.
Note: specific model-version recommendations, such as minimum GPT or Gemini versions, will be added here once confirmed through testing.
Privacy
Your code is sent only to the provider you configure, and only for the actions that require a model call: ship-readiness checks and cleanup. Security scanning and file splitting never call a provider at all. Nothing passes through Zeno’s servers. See Config Files for where credentials are stored locally.