3 min read

Connect Claude (MCP)

Account-based marketing for AI agents. Claude can call abm.dev directly — connect once, then ask Claude to enrich a person or a company.

Connect once. Then ask Claude to enrich a person or a company, and it returns eighty-nine canonical fields and forty signals — each one carrying its source, a confidence score, and a selection_reason.

Connect the MCP server

The MCP server lives at one URL:

Text

https://mcp.abm.dev/mcp

In Claude (web or desktop):

Open Settings → Connectors.
Choose Custom.
Paste https://mcp.abm.dev/mcp.

That’s it. Claude now sees abm.dev’s enrichment tools and can run enrichments inside a conversation.

Tip

The server exposes abm.dev’s enrichment tools in Claude. We don’t list the exact tool names here — they can change. For the current set, see /for-agents.

What you get back

Ask for a person, a company, or both. The response is resolved from ten data sources — LinkedIn, Companies House, Perplexity, Tavily, Hunter, and others — into one answer.

Every value travels with three things:

Source. Which provider it came from.
Confidence. A score from 0 to 1. Higher means more agreement across sources.
Selection_reason. Why that value was chosen over the alternatives.

No per-source bills. No per-field charges. No silent fallbacks. A value is cited, or it isn’t returned. Nothing is invented.

To go deeper on the model, see Confidence scores, Data sources, and Canonical fields.

Other agents (Cursor, Claude Code, Windsurf)

These coding agents don’t take a connector URL. They take a prompt. Paste this, and let the agent do the setup:

Tip

Read https://abm.dev/llms.txt for context. Then request an API key, save it to a project .env, run one enrichment as a smoke test, and report the result — including which source cited each field.

The agent reads the docs, gets a key, stores it, and proves the connection with a real call. For the full agent walkthrough, see /for-agents.

Scripts and terminals

No agent in the loop? Call the REST API directly.

Authenticate with a header. Send JSON.

Bash

curl https://api.abm.dev/api/v2/enrichments \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"targets":[{"type":"person","email":"[email protected]"}]}'

Each target also accepts a LinkedIn URL or a company domain.

Note

Send your key as either x-api-key (shown here) or Authorization: Bearer on POST /api/v2/enrichments. The full, live contract is at the API reference.

Read a single enrichment back by id:

Text

GET https://api.abm.dev/api/v2/enrichments/{id}

Full request and response shapes live in the API reference.

Real-time and batch

For agent loops, enrichments stream in real time over SSE — results arrive as they resolve. For batch work, webhooks fire when a job finishes. Setup specifics — endpoints, payloads — are covered in Jobs.

What it costs

Per enrichment. No subscription. Credits never expire. All ten sources are included in the price.

From about €0.29 per enrichment. Packs:

Pack	Price
30	€2.89
100	€9.29
500 (best value)	€36.99
2,000	€119.99

The playground is free. New accounts get launch credits with code LAUNCHCODES.

See pricing, or try it in the playground.

Next steps

Getting started — your first enrichment, end to end.
Claude connector guide — the connector in detail.
/for-agents — the agent-first overview.
Sign up — get a key.

Who runs it

abm.dev is operated by Foxley Farm Operations Ltd (UK Company No. 16392009), Oakham, United Kingdom. GDPR-compliant.

Stuck? Open a support ticket.