Docs

Use CallAnswer with your AI agent

Everything CallAnswer does for you in the dashboard — see captured leads and transcripts, check the calendar, book or cancel an appointment, read or update business hours — your own AI agent can do too. There are three doors, all opened by one API key from Settings → Agent access: a remote MCP server for Claude, ChatGPT/Codex, Cursor, VS Code and any other MCP client; a REST API with an OpenAPI document; and the callanswer CLI. The agent acts as you, reaching only your business, and cannot send an SMS, place a call, or touch billing.

1. Get an API key

Sign in and open Settings → Agent access. Name the key after the thing that will use it (“Claude Code on my laptop”) and pick a scope: read for listing calls, appointments and hours; write to also book/cancel appointments and change hours. The key is shown once. Put it in an environment variable — every snippet below reads CALLANSWER_API_KEY — rather than pasting it into a config file that gets committed.

2. Claude Code, claude.ai and Claude Desktop

Claude Code, from a terminal (the key is sent as a header on every request):

claude mcp add --transport http callanswer https://callanswer.net/mcp \
  --header "Authorization: Bearer $CALLANSWER_API_KEY"

Or add it to a project's .mcp.json so the whole team gets it; Claude Code expands ${CALLANSWER_API_KEY} from the environment:

{
  "mcpServers": {
    "callanswer": {
      "type": "http",
      "url": "https://callanswer.net/mcp",
      "headers": { "Authorization": "Bearer ${CALLANSWER_API_KEY}" }
    }
  }
}

claude.ai and Claude Desktop: Customize → Connectors → Add custom connector. Name it CallAnswer, set the URL to https://callanswer.net/mcp, choose “No sign in”, and under Request headers add Authorization with the value Bearer <your key>. Claude will ask before running cancel_appointment.

3. Cursor and VS Code

Cursor — ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (one project):

{
  "mcpServers": {
    "callanswer": {
      "url": "https://callanswer.net/mcp",
      "headers": { "Authorization": "Bearer ${env:CALLANSWER_API_KEY}" }
    }
  }
}

VS Code — .vscode/mcp.json; the input prompt keeps the key out of the file:

{
  "servers": {
    "callanswer": {
      "type": "http",
      "url": "https://callanswer.net/mcp",
      "headers": { "Authorization": "Bearer ${input:callanswer-key}" }
    }
  },
  "inputs": [{ "id": "callanswer-key", "type": "promptString", "password": true, "description": "CallAnswer API key" }]
}

4. ChatGPT and Codex

Codex CLI and the ChatGPT desktop app share one MCP configuration:

[mcp_servers.callanswer]
url = "https://callanswer.net/mcp"
bearer_token_env_var = "CALLANSWER_API_KEY"

ChatGPT's in-app custom connectors currently require OAuth or no authentication, and cannot send a fixed header. Until CallAnswer has an OAuth authorization server, use Codex or a direct API key with another client.

5. The CLI

Node 18 or newer. No install needed; npx fetches the callanswer package. It talks to the same API with the same key and the same limits, and prints tables for people or JSON for scripts:

# one-time: paste your key when prompted (stored in ~/.config/callanswer/config.json, mode 600)
npx callanswer login

npx callanswer whoami
npx callanswer calls list
npx callanswer calls get <id>
npx callanswer appointments list
npx callanswer appointments create <leadId> "2026-10-15T14:00:00-04:00"
npx callanswer appointments cancel <id> --yes            # refuses without --yes
npx callanswer hours get
npx callanswer hours set hours.json --yes

# any command: --json for machine-readable output; CALLANSWER_API_KEY overrides the saved key

6. The tools

What any MCP client sees from tools/list. “Read-only” and “destructive” are the server's own annotations; Claude and other clients use them to decide when to ask you first.

Try the endpoint by hand — initialize and tools/list need no key:

curl -s -X POST https://callanswer.net/mcp \
  -H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18","capabilities":{},"clientInfo":{"name":"curl","version":"0"}}}'

7. REST API and OpenAPI

The MCP tools and the CLI are thin layers over /api/v1. The full description is at /openapi.json (OpenAPI 3.1), which any code generator or agent framework can read directly.

curl -s "https://callanswer.net/api/v1/calls?status=NEW&limit=10" \
  -H "Authorization: Bearer $CALLANSWER_API_KEY"

curl -s -X POST https://callanswer.net/api/v1/appointments \
  -H "Authorization: Bearer $CALLANSWER_API_KEY" -H "Content-Type: application/json" \
  -d '{"leadId":"<id>","date":"2026-10-15T14:00:00-04:00","confirm":true}'

Endpoints: GET /api/v1/me, GET /api/v1/calls?q=&status=, GET /api/v1/calls/{id}, GET|POST /api/v1/appointments, POST /api/v1/appointments/{id}/cancel, GET|PATCH /api/v1/business-hours.

8. Safety, limits and logging

• An API key acts as you. An agent reaches only your business — never another customer's.
• Nothing here can send an SMS, place a call, or touch billing.
• create_appointment, cancel_appointment and update_business_hours all need confirm: true; the first call only previews the change. cancel_appointment is additionally destructive. The CLI needs --yes.
• Rate limit: 300 agent calls per key per hour.
• Every authenticated call is logged (tool, key, outcome, duration) and shown back to you at GET /api/v1/me / the whoami tool; rows are kept 30 days.
• Keys are stored hashed. Revoke one in Settings and it stops working immediately. Deleting your account deletes its keys.

Reviewed 2026-10-02. Protocol facts checked against modelcontextprotocol.io on that date.