Integrations
MCP Server
Overview
Conturs MCP Server lets you score leads directly inside AI assistants through natural conversation — no code or API calls needed.
The scoring logic is identical to the Scoring API. Same 13 factors, same weights, same results.
Available Tools
The MCP server exposes two tools:
get_config — fetches your scoring configuration (weights for each factor). Your API key is read from the connection config automatically.
calculate_score — scores a lead using contact data and your config weights. Returns a 0–100 score, label (Hot / Warm / Cool / Cold), and factor breakdown.
For all supported fields and response format, see API Reference.
Setup — Claude Desktop
Open Settings → Developer → Edit Config to open claude_desktop_config.json and add:
Restart Claude Desktop after saving.
Usage
Once connected, just ask in natural language:
"Get my scoring config"
"Score this lead: jane@startup.io, CEO at Startup Inc, qualified, deal amount $50k"
"Score these 3 leads and tell me which one to prioritize"
Claude handles the tool calls automatically.
API Key
Get your API key from Dashboard → API. The same key works for both the MCP server and the REST API.
