API introduction
Leadveyor turns incoming messages into scored leads. Send it emails, form submissions or any text; it filters junk, extracts who the customer is and what they want, merges repeat messages from the same person, and scores every lead by criteria you define — with the reason for every point.
How it works
- A message arrives — through the API, an email address Leadveyor gives you, or a web form URL. Each one becomes an inbound in the processing queue.
- Junk is filtered — spam, newsletters, sales pitches and auto-replies are set aside with a reason. Your own rules and blocked domains apply first.
- Leads are extracted — contact details, a short summary, the address with distance and drive time from your business, and the fields you ask for (budget, date, guest count…). One message can contain several leads.
- Repeat messages are merged — the same person writing again within 30 days updates their lead instead of creating a new one. Every lead is linked to its contact.
- Leads are scored — your profile turns facts and answers into a score from 0 to 100 and a tier (for example hot / warm / cold). The number is always computed by code, never guessed by the model, and every AI answer quotes the message as evidence.
- You get the result — in the API response, through webhooks, or by polling leads and events. When a deal is won or lost, send the outcome back.
The base URL of the API is https://core.leadveyor.com. All requests and responses are JSON. The machine-readable spec is at /openapi.json, and a plain-text version of these docs for AI agents is at /llms.txt. To give an AI assistant (Claude, Cursor, your own agent) access to the same operations, connect it to the MCP server at https://core.leadveyor.com/mcp with an API key.