{
  "markdown": "# ChatFlow MCP\n\n**Connect your AI agent to your WhatsApp CRM.**\n\n[ChatFlow](https://chatflow.biz) is a WhatsApp/Telegram/Instagram CRM for small\nteams — a shared inbox, a drag-and-drop sales pipeline, AI auto-replies, and\nbroadcast campaigns. This is its hosted **MCP (Model Context Protocol)** server:\npoint Claude, Claude Code, Cursor, or any MCP-compatible agent at your workspace\nand it can search contacts, read conversations, manage your pipeline, and — when\nyou allow it — send messages.\n\n- **Endpoint:** `https://valuable-sockeye-14.eu-west-1.convex.site/mcp`\n- **Transport:** Streamable HTTP (stateless)\n- **Auth:** `Authorization: Bearer <API key>` — create one in ChatFlow →\n  Settings → AI agents (MCP)\n\n## Quick start\n\n1. In ChatFlow, open **Settings → AI agents (MCP)** (workspace owner) and click\n   **Create API key**. Copy it — it's shown once. Optionally give it an expiry.\n2. Add the server to your client:\n\n**Claude Code**\n```bash\nclaude mcp add --transport http chatflow \\\n  https://valuable-sockeye-14.eu-west-1.convex.site/mcp \\\n  --header \"Authorization: Bearer cfk_YOUR_KEY\"\n```\n\n**Claude.ai / Claude Desktop** — Settings → Connectors → Add custom connector →\npaste the endpoint URL and set the `Authorization: Bearer cfk_YOUR_KEY` header.\n\n**Cursor** — `.cursor/mcp.json`:\n```json\n{\n  \"mcpServers\": {\n    \"chatflow\": {\n      \"url\": \"https://valuable-sockeye-14.eu-west-1.convex.site/mcp\",\n      \"headers\": { \"Authorization\": \"Bearer cfk_YOUR_KEY\" }\n    }\n  }\n}\n```\n\nFull setup and troubleshooting: [docs/setup.md](docs/setup.md).\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `search_contacts` | Find contacts by name/phone substring |\n| `list_conversations` | Recent chats with previews, unread counts, CRM fields |\n| `get_conversation` | One chat: contact, notes, deal value, recent messages |\n| `list_pipeline` | Kanban columns with the leads in each stage |\n| `move_conversation` | Move a lead to a pipeline column |\n| `update_lead` | Set notes / deal value on a conversation |\n| `create_contact` | Create a WhatsApp contact + conversation from a phone number |\n| `send_message` | Send a text to the customer (opt-in — see below) |\n| `get_usage` | Subscription tier, limits, current monthly usage |\n\nRead tools are annotated read-only; `send_message` is annotated as a\nworld-affecting write so well-behaved clients confirm before calling it.\n\n## Sending is opt-in\n\n`send_message` messages a **real customer**, so it's off by default. The\nworkspace owner enables it with the **\"Allow agents to send messages\"** toggle\nin Settings. While off, agents can read everything and draft replies, but every\nsend is rejected with a message explaining the toggle. When on, agent sends\ncount toward your monthly quota, appear in the inbox with the AI badge, pause\nthe auto-reply assistant on that chat (so two AIs never answer at once), and\nsupport an `idempotencyKey` so a retry can't double-send.\n\n**Prompt-injection note:** message text and notes an agent reads are written by\nyour customers. Keep the send toggle off unless you trust the agent driving it.\n\n## Links\n\n- Product: <https://chatflow.biz>\n- Setup guide: [docs/setup.md](docs/setup.md)\n",
  "bytes": 3169,
  "sha": "50d041ebfbabf159cba3e449cdaa5c2a56d4028042ed84143ffedfbdd88857db",
  "repo_slug": "yersat/chatflow-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_yersat_chatflow_abe71a39/readme"
}