Anthropic Claude

Status: Live on conc-exe.xyz · Executive Lounge model picker + API

API base: https://api.anthropic.com/v1/messages

Operator keys: platform.claude.com/settings/keys ↗

Concierge adapter: backend/concierge-api/concierge-anthropic.ts

1. Operator sets ANTHROPIC_API_KEY (sk-ant-*) on the Concierge deployment.
2. Client selects claude-sonnet-4-6 or claude-haiku-4-5 in Lounge, or sends agentModel in the chat body.
3. Concierge maps the id to Anthropic model ids and calls POST /v1/messages.
4. x402 USDC (or Token Pay) settles the Concierge route — consumers do not need an Anthropic key when calling production.
5. On API error or missing operator key, Concierge falls back to Gemini (modelFallback: true).

Image, enhance, and trading-plan modes always use Gemini. Claude models apply to mode: "chat" only. The Anthropic account owner name is never exposed to end users.

Aliases accepted in the API: claude, sonnet, haiku. Registry: concierge-llm-models.ts.

See repo docs/configuration.md and .env.example.

pay --sandbox curl https://conc-exe.xyz/api/concierge \
  -H "Content-Type: application/json" \
  -d '{
    "mode": "chat",
    "message": "Summarize BTC vs ETH momentum this week.",
    "history": [],
    "agentModel": "claude-sonnet-4-6"
  }'

Swap agentModel to claude-haiku-4-5 for the fast tier. Expect HTML in reply and modelUsed reflecting the upstream model when Anthropic succeeds.

• Concierge Chat API — agentModel and response shape
• HYRE Gateway — alternate optional chat models
• OpenAI GPT-5.6 — alternate optional chat models
• MCP Registry — Claude Desktop connects at /api/mcp
• pay.sh — pay claude for agents calling Concierge intel
• Anthropic Platform docs ↗