SparteraConnect

SparteraConnect provides managed MCP (Model Context Protocol) servers as a service, enabling AI agents to query your Spartera analytics assets using natural language — no custom integration code required.

Price: $100/month per organization

What Is MCP?

Model Context Protocol (MCP) is an open standard that lets AI assistants like Claude, ChatGPT, and custom LLMs call external tools and data sources directly during a conversation. SparteraConnect hosts and manages the MCP server infrastructure so your AI agents can access Spartera analytics without any backend work on your end.

What SparteraConnect Enables

Without SparteraConnect, an AI agent has no way to query live analytics. With SparteraConnect, an agent can:

Call any of your Spartera analytics assets by name
Ask questions in natural language and receive computed results
Chain multiple analytics calls in a single conversation
Access marketplace assets your organization has purchased

Example interaction:

User to Claude: "What's our customer churn risk this month compared to last month?"

Claude (via SparteraConnect): Calls your "Monthly Churn Risk Score" Spartera asset, retrieves both periods, compares them, and responds with a summary.

Getting Started

Step 1: Subscribe to SparteraConnect

Dashboard → SparteraConnect → Enable

Your managed MCP server is provisioned within minutes at:

https://mcp.spartera.com/{your-company-id}

Step 2: Connect Your AI Agent

Claude (via claude.ai or API)

Add your SparteraConnect server to your Claude MCP configuration:

{
  "mcp_servers": [
    {
      "type": "url",
      "url": "https://mcp.spartera.com/{your-company-id}",
      "name": "spartera",
      "auth": {
        "type": "api_key",
        "key": "your-spartera-api-key"
      }
    }
  ]
}

Custom LLM Application (via Anthropic API)

const response = await fetch("https://api.anthropic.com/v1/messages", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    model: "claude-sonnet-4-20250514",
    max_tokens: 1000,
    messages: [
      { role: "user", content: "What's our average deal size this quarter?" }
    ],
    mcp_servers: [
      {
        type: "url",
        url: "https://mcp.spartera.com/your-company-id",
        name: "spartera-analytics"
      }
    ]
  })
});

Step 3: Give Your Agent Access to Assets

By default, your MCP server exposes all assets your organization owns or has purchased. You can scope access:

Dashboard → SparteraConnect → Asset Permissions

How It Works Under the Hood

User asks AI agent a question
        │
        ▼
AI agent identifies relevant Spartera asset via MCP tool discovery
        │
        ▼
MCP server authenticates with Spartera API using your org credentials
        │
        ▼
Spartera executes asset SQL inside your database (zero data movement)
        │
        ▼
Result returned to MCP server → to AI agent → to user

Your data still never leaves your infrastructure. SparteraConnect is purely an orchestration layer.

Available MCP Tools

SparteraConnect automatically exposes these tools to connected AI agents:

Tool	Description
`list_assets`	List all accessible analytics assets
`execute_asset`	Run an asset and return results
`get_asset_details`	Get metadata, description, and schema for an asset
`search_marketplace`	Search the Spartera marketplace for relevant assets
`get_usage_stats`	Retrieve execution history and credit usage

Use Cases

Internal Analytics Copilot

Give your team a Slack bot or internal assistant that can answer data questions by calling your proprietary analytics assets — without exposing raw database access.

Customer-Facing AI

Embed Spartera analytics into your product's AI assistant. When a customer asks "how is my campaign performing?", your AI calls the relevant Spartera asset and responds with real data.

Automated Reporting

Set up scheduled AI workflows that pull analytics, generate narratives, and distribute reports — all via MCP with no manual intervention.

AI Agent Pipelines

Chain Spartera analytics calls in multi-step AI agent workflows — e.g., "get churn risk → identify at-risk segments → generate outreach copy."

Pricing

Plan	Price	Included
SparteraConnect	$100/month	Unlimited MCP queries, all your assets, standard support
Enterprise	Custom	Dedicated server, SLA, SSO, priority support

MCP queries do not consume extra credits — asset execution costs apply as normal per your existing asset pricing.

FAQ

Does SparteraConnect work with any AI?
Any AI that supports MCP can use SparteraConnect. This includes Claude, and any custom application built with the Anthropic API, OpenAI API, or open-source models with MCP support.

Is my data secure?
Yes. SparteraConnect is an orchestration layer only. It passes API calls through to your Spartera assets — it never stores, logs, or caches your actual data.

Can I restrict which assets are accessible via MCP?
Yes. Use the Asset Permissions panel in the SparteraConnect dashboard to allow/deny specific assets.

What happens if an asset fails mid-conversation?
The MCP server returns a structured error to the AI agent, which can handle it gracefully in its response.