Works with Claude, Cursor and other AI assistants
Ask about your leads in plain words.
Connect Leadveyor to the AI assistant you already use. Ask who to call first, why a lead scored low, or change your scoring rules — just by typing. No reports to build, no settings to dig through.
Which hot leads came in this week that nobody has answered yet?
Looked at this week’s leads in Leadveyor
Three hot leads are still waiting:
- 86
Sarah Mitchell
Wedding · 120 guests · 14 March · Paddington
- 78
James Lee
Corporate lunch · 60 people · next Friday
- 72
Priya Natarajan
50th birthday · 80 guests · April
Call Sarah first: the date and venue are confirmed, the venue is 6 km from your kitchen, and she wrote “we’d love to book this week”.
What you can ask
Everything you’d open the dashboard for.
Your assistant reads your leads, explains every score and makes changes for you — in the same chat where you already work.
Morning briefing
“What came in overnight? Anything urgent?”
New leads ranked by score, with a one-line summary of each and who to call first.
Explain a score
“Why is the Lee enquiry only warm?”
Every reason, with the exact words from the email: “no date yet”, “budget not mentioned”.
Record results
“The Mitchell wedding is booked, $18,500. The Smith party went elsewhere on price.”
Marks one won and one lost with the reason, so Leadveyor learns what a good lead looks like.
Change the rules
“Events outside Sydney aren’t worth it for us. Score them lower.”
Drafts the rule, tests it on past emails, shows the difference, and waits for your yes.
Check the spam folder
“Was anything real caught as spam this week?”
Lists what was filtered out and why, and puts a real enquiry back in the queue.
Set things up
“Give me an email address for our website enquiries.”
Creates it on the spot. Add a webhook to your CRM or a new workspace the same way.
Change the rules by asking
Teach it what a good lead is. In a sentence.
Your business changes; your scoring should too. Tell your assistant what matters now, and it rewrites the rules for you.
Tested before it’s live
New rules are tried on your past enquiries first. You see which leads move up or down.
Nothing changes without you
The new rules switch on only when you say yes. Until then, scoring works as before.
One step back
Don’t like the result? Ask to go back to the previous rules.
Corporate events are our best work now. Score them higher — but test it on last week’s emails first.
Read your scoring rules
Wrote a new rule: “Corporate event” +20 points
Tested on 12 emails from last week — nothing changed yet
| Enquiry | Now | With the new rule |
|---|---|---|
| Office lunch, 60 people | 58 Warm | 78 Hot |
| Christmas party, 120 staff | 55 Warm | 75 Hot |
| Birthday, 30 guests | 44 Warm | 44 Warm |
Corporate enquiries move up to hot; everything else stays as it was. Shall I switch the new rules on?
How to connect
Three steps. No developer needed.
- 1
Create an access key
In Leadveyor, open API keys and create one for your assistant. Choose “read” to let it look, or “write” to let it make changes too.
- 2
Add Leadveyor to your assistant
Paste the Leadveyor address and the key into your assistant’s settings. It takes about two minutes; our guide shows each app step by step.
- 3
Ask
Type your question the way you’d ask a colleague. The assistant finds the answer in Leadveyor and shows it to you.
Works with
- Claude Desktop
- Claude Code
- Cursor
- VS Code
- Claude API
- Your own AI agent
You stay in control
Helpful, not in charge.
Your assistant works for you, with the permissions you give it — nothing more.
Look-only keys
A “read” key lets the assistant look at leads and explain them, but not change anything.
Asks before big changes
Switching on new scoring rules or deleting a setting needs your confirmation in the chat.
Test first, then live
Rule changes are tried on past enquiries before they touch a single new lead.
Leads can’t be deleted
Your assistant can’t delete leads or create new access keys. Those stay in your hands.
Same rules as your team
The assistant sees and does only what its key allows — exactly like any other connection.
Off in one click
Revoke the key in Leadveyor and the assistant is disconnected immediately.
For developers and agencies
Building your own AI agent?
Leadveyor speaks MCP, the open standard AI agents use to work with other apps. 21 tools cover leads, outcomes, scoring rules, sources, webhooks and workspaces — with the same keys and permissions as our REST API.
claude mcp add --transport http leadveyor \
https://core.leadveyor.com/mcp \
--header "Authorization: Bearer lv_live_…"Questions, answered simply.
- What is MCP?
- The Model Context Protocol — an open standard that lets AI assistants use other apps on your behalf. Leadveyor supports it, so assistants that speak MCP can read your leads and change settings when you ask.
- Which assistants work?
- Today: Claude Desktop, Claude Code, Cursor, VS Code, AI agents built on the Claude API, and any other app that accepts an MCP server address with an access key. The web versions of Claude and ChatGPT connect with a sign-in button instead of a key, so they can’t connect yet.
- Do I still need the Leadveyor console?
- For first setup and to create the access key, yes. After that, you can do most everyday things from your assistant: review leads, record results, adjust scoring.
- Can my assistant see customer details?
- Yes — names, emails and messages of your leads, the same as your team sees. Connect only an assistant your business already trusts with customer data, and use a look-only key if it doesn’t need to change anything.
- Can I try it without touching real leads?
- Yes. Create a test key: everything the assistant creates with it is kept apart from your live leads. Scoring a sample message also works without saving anything.
Stop sorting. Start calling the right people first.
Free while in beta. We’re letting businesses in a few at a time and we’ll help you set up your first rules.