Docs / Routing

Per-client config snippets

Point your AI client at TokenOne®. The Agent Switcher modal in the TokenOne Delivery® Workspace generates these automatically · this page has the raw text so you can drop them into CI / onboarding scripts.

Claude Code

# ~/.claude/settings.json OR env vars
export ANTHROPIC_BASE_URL="https://api.tokenone.io/v1/anthropic-proxy"
export ANTHROPIC_API_KEY="to_live_..."

# First session on a project · auto-injects the handoff via MCP:
claude /mcp tool terminal_handoff projectId=<uuid>

Cursor

// Cursor settings → Models → Override OpenAI Base URL
{
  "openai.baseUrl": "https://api.tokenone.io/v1/openai-proxy",
  "openai.apiKey": "to_live_..."
}

Windsurf

// Windsurf settings → BYOK → Custom provider
{
  "provider": "openai-compat",
  "baseUrl": "https://api.tokenone.io/v1/openai-proxy",
  "apiKey": "to_live_..."
}

Cline (VS Code)

// VS Code settings.json → Cline → Custom API Provider
{
  "cline.apiProvider": "openai-compatible",
  "cline.openAiCompatible.baseUrl": "https://api.tokenone.io/v1/openai-proxy",
  "cline.openAiCompatible.apiKey": "to_live_..."
}

Continue

// ~/.continue/config.json
{
  "models": [
    {
      "title": "TokenOne® TokenOne Delivery",
      "provider": "openai",
      "apiBase": "https://api.tokenone.io/v1/openai-proxy",
      "apiKey": "to_live_...",
      "model": "claude-sonnet-4"
    }
  ]
}

Aider

# ~/.aider.conf.yml
openai-api-base: https://api.tokenone.io/v1/openai-proxy
openai-api-key: to_live_...

# or env vars:
export OPENAI_API_BASE="https://api.tokenone.io/v1/openai-proxy"
export OPENAI_API_KEY="to_live_..."

Generic Gemini client

// Any Gemini SDK supporting custom baseUrl:
{
  "baseUrl": "https://api.tokenone.io/v1/gemini-proxy",
  "apiKey": "to_live_..."
}

Test any of these with a single curl against the matching /health endpoint before wiring · the Agent Switcher has a one-click test button that does this for you.