Adding API Keys
Add your own provider API keys (BYOK) so Clevername can route governed agent traffic through Anthropic, OpenAI, Google, Azure OpenAI, or any LiteLLM-supported model. BYOK is included on every tier — Free, Pro, Team, and Enterprise.
Why BYOK?
Clevername is fully BYOK — there is no managed-LLM tier and no credit system on any plan. You add API keys from the providers you already use (Anthropic, OpenAI, Google, Azure OpenAI, or any LiteLLM-supported model) and Clevername routes governed agent traffic through them. You pay providers directly at their published rates, keep full control over your usage limits, and your keys are never shared with other users. This applies on every tier including Free.
How keys are stored
Your API keys are encrypted and stored in GCP Secret Manager. After you submit a key, Clevername stores only a masked hint (e.g., sk-...7x4Q) in the database for display purposes. The raw key value is never stored in the application database.
Supported providers
- Anthropic — Claude Opus 4.6, Sonnet 4.6, Haiku 4.5
- OpenAI — GPT-5.2, GPT-4o, o3, o4-mini
- Google — Gemini 3.1 Pro, Gemini 2.5 Flash
Navigate to Settings > Keys
From the dashboard, click Settings in the sidebar, then select the Keys tab. This page shows all your configured API keys and lets you add new ones.
Enable MFA (if not already active)
Managing API keys requires multi-factor authentication. If you haven't set up MFA yet, you'll be prompted to do so before you can add or modify keys. Follow the MFA setup flow to configure an authenticator app.
Click Add Key and select a provider
Click the Add Key button. Select the provider you want to configure from the dropdown (Anthropic, OpenAI, or Google). You can store multiple keys per provider using labels.
Paste your API key
Paste your API key from the provider's dashboard. You can get your keys from:
- Anthropic:
console.anthropic.com→ API Keys - OpenAI:
platform.openai.com→ API Keys - Google:
aistudio.google.com→ API Keys
Save and verify
Click Save. Clevername stores your key in GCP Secret Manager and validates it with a lightweight test call to the provider. If the key is invalid, you'll see an error message and the key is not saved. Once saved, the provider's models become selectable when configuring agents in AI Company → Agents, and key health is re-validated automatically every six hours.