MCP Integration Guide

Step-by-step instructions to connect your AI coding tools to Nubiq via the Model Context Protocol.

Getting Started

Create an API Key

Sign in to your Nubiq dashboard
Navigate to Settings → API Keys
Click Create API Key
Select scopes: at minimum, enable the tools you need (e.g., conversations:read)
Copy the key and store it securely

Choose Your Client

Pick your AI coding tool below and follow the setup instructions.

Client Setup

🧠Claude Code

Works immediately after running the command. Claude Code will discover all 30 tools automatically.

Terminal

claude mcp add nubiq --transport streamable-http \
  --url https://hub.nubiq.ai/api/mcp/ \
  --header "Authorization: Bearer YOUR_API_KEY"

⌨️Cursor

Create or edit .cursor/mcp.json in your project root. Restart Cursor to load the server.

.cursor/mcp.json

{
  "mcpServers": {
    "nubiq": {
      "transport": "streamable-http",
      "url": "https://hub.nubiq.ai/api/mcp/",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

🏄Windsurf

Add to your Windsurf MCP configuration file. Restart the editor to apply.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "nubiq": {
      "serverUrl": "https://hub.nubiq.ai/api/mcp/",
      "transport": "streamable-http",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

▶️Continue.dev

Add the nubiq entry to the mcpServers array in your Continue config.

~/.continue/config.json (mcpServers section)

{
  "mcpServers": [
    {
      "name": "nubiq",
      "transport": {
        "type": "streamable-http",
        "url": "https://hub.nubiq.ai/api/mcp/",
        "headers": {
          "Authorization": "Bearer YOUR_API_KEY"
        }
      }
    }
  ]
}

🔧Cline

Open VS Code settings, search for Cline MCP, and add the server configuration.

VS Code Settings > Cline > MCP Servers

{
  "mcpServers": {
    "nubiq": {
      "transport": "streamable-http",
      "url": "https://hub.nubiq.ai/api/mcp/",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Tool Reference

Detailed input schemas for the most commonly used tools. All tools follow the JSON-RPC 2.0 protocol.

conversations.list

List conversations with optional filters.

Input Schema

{
  "status": "open" | "handoff" | "closed",
  "assigned_to": "agent-uuid",
  "limit": 20,
  "cursor": "next-page-cursor"
}

conversations.reply

Send a reply to a conversation as an agent.

Input Schema

{
  "thread_id": "conv-uuid",
  "content": "Your reply message",
  "internal": false
}

documents.search

Semantic search over the knowledge base.

Input Schema

{
  "query": "refund policy",
  "top_k": 5,
  "threshold": 0.7
}

documents.upload

Upload a document to the knowledge base.

Input Schema

{
  "name": "FAQ.pdf",
  "content_type": "application/pdf",
  "content_base64": "base64-encoded-content"
}

tickets.create

Create a support ticket.

Input Schema

{
  "subject": "Billing issue",
  "description": "Customer cannot access...",
  "priority": "high",
  "contact_id": "contact-uuid"
}

analytics.overview

Get KPI summary for the tenant.

Input Schema

{
  "period": "7d" | "30d" | "90d"
}

← View all 30 tools on the Developers page

Example Workflows

Upload a document and search it

documents.upload({ name: "FAQ.pdf", content_base64: "..." })
// Wait for indexing to complete
documents.search({ query: "How to reset password?", top_k: 3 })

Monitor conversations and auto-tag

conversations.list({ status: "open" })
// For each conversation, analyze the latest messages
conversations.get({ thread_id: "..." })
// Based on content, update status or assign
conversations.assign({ thread_id: "...", agent_id: "..." })

Create tickets from conversations

conversations.list({ status: "open" })
// Identify conversations that need tickets
tickets.create({ subject: "...", description: "...", contact_id: "..." })
conversations.update_status({ thread_id: "...", status: "closed" })