Sponsored by Deepsite.site

abm.dev

创建者
abm-dev-gita month ago
The account-based marketing API for AI agents. One key, one schema: find the right accounts and people, enrich them into cited multi-source fields, and write records and outreach straight to your CRM. Every value carries its source and a confidence score.
概览

abm.dev

abm.dev MCP server

The account-based marketing API for AI agents.
Search, Enrich, and Create across B2B people and companies — every field cited with a source and confidence score. One key, ten sources, no per-field bills.

npm MIT MCP Smithery


What your agents can do

abm.dev is the account-based marketing API for AI agents. One key, one schema, four verbs:

  • Search — find companies and people by criteria. search_companies, search_contacts.
  • Source — hand it a target account, get the right buying-committee people back. Async briefs: source, get_source.
  • Enrich — cited, multi-source fields on any person or company. create_enrichment, get_enrichment_fields, get_enrichment_sources.
  • Create — write CRM records and stage outreach. create_company, create_contact, create_outreach; queue_linkedin_post stages a post for human approval — it never auto-publishes.

Every value carries its source and a confidence score (0–1). A value is cited or it is not returned. No fabricated facts.

What you get

  • Eighty-nine canonical fields per record — forty-three on the person, forty-six on the company.
  • Forty signals on every prospect.
  • Ten sources, resolved into one response — LinkedIn, Companies House, Perplexity, Tavily, Hunter, and others.
  • A citation on every valuesource and confidence (0–1), with the winning source that produced it.

No per-source bills. No per-field charges. No silent fallbacks.


Connect

Claude (custom connector)

  1. Open Settings → Connectors → Custom.
  2. Paste the server URL: https://mcp.abm.dev/mcp
  3. Complete the one-time sign-in when prompted.

Then ask, in plain language: "Find the VP of Marketing at Acme, then enrich them."

Claude Code (plugin)

One install wires both the connector and the abm.dev Agent Skill — no URL pasting:

/plugin marketplace add abm-dev-git/abm-mcp
/plugin install abm-dev@abm-dev

OAuth sign-in happens on first tool use. See claude-plugin/.

Cursor / Claude Code / Windsurf

These agents drive the REST API directly. Paste this prompt:

Read https://abm.dev/llms.txt. Request an API key, save it to .env as ABM_API_KEY, then run a test enrichment against https://api.abm.dev/v1/enrichments. Report the result with its citations.

Script and terminal agents

Call the REST API directly:

curl https://api.abm.dev/v1/enrichments \
  -H "x-api-key: $ABM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"type":"person","input":{"email":"jane@acme.com"}}'

input also accepts a LinkedIn URL or a company domain.


Install (npx)

Prefer a local command over a remote URL? The @abmdev/mcp package is a thin bridge: npx -y @abmdev/mcp launches a local STDIO MCP server that proxies to the hosted server at https://mcp.abm.dev/mcp, handling the OAuth 2.1 PKCE sign-in and token caching for you. It defines no tools of its own — it exposes whatever the hosted server serves.

Add it to any MCP client that supports command/args (Claude Desktop, Cursor, Windsurf, Claude Code, and others):

{
  "mcpServers": {
    "abm": {
      "command": "npx",
      "args": ["-y", "@abmdev/mcp"]
    }
  }
}

Generic config snippet (for clients that take a bare command):

{ "command": "npx", "args": ["-y", "@abmdev/mcp"] }

On first run a browser window opens for the one-time sign-in; the token is then cached for subsequent runs. Point the bridge at a different endpoint by setting the ABM_MCP_URL environment variable.

Either path reaches the same hosted server. Use the custom connector (remote URL) above if your client supports it directly; use npx for clients that only speak local command/args.


Tools

Grouped by verb — every field comes back with its source and confidence (0–1). Real-time SSE streaming for agent loops; webhooks fire when batch jobs finish.

Searchsearch_companies, search_contacts: find accounts and people by criteria.

Sourcesource, get_source: give it an account, get the right people back (async brief).

Enrichcreate_enrichment, get_enrichment, get_enrichment_fields, get_enrichment_sources: cited, multi-source fields on a person (from an email or LinkedIn URL) or a company (from a domain).

Createcreate_company, create_contact, create_outreach, queue_linkedin_post: write CRM records and stage outreach. Staged posts wait for human approval; nothing publishes automatically.


Pricing

Per-enrichment. No subscription. Credits never expire. All ten sources included — no add-ons.

From about €0.29 per enrichment. Packs: 30 credits €2.89, 100 €9.29, 500 €36.99 (best value), 2,000 €119.99. The playground is free. Free launch credits with code LAUNCHCODES.

See https://abm.dev/#pricing.



Registry reference

This repo is the public description of a hosted MCP server. The server runs at https://mcp.abm.dev/mcp; this repo is not its source. See server.json for the machine-readable manifest.

Support

Questions or a stuck enrichment? Open a ticket: https://abm.dev

Operator

Foxley Farm Operations Ltd · UK Company No. 16392009 · Oakham, United Kingdom. GDPR-compliant.

服务器配置

{
  "mcpServers": {
    "abm": {
      "command": "npx",
      "args": [
        "-y",
        "@abmdev/mcp"
      ]
    }
  }
}
推荐的 MCP Server
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
ChatWiseThe second fastest AI chatbot™
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
DeepChatYour AI Partner on Desktop
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
WindsurfThe new purpose-built IDE to harness magic
Tavily Mcp
Playwright McpPlaywright MCP server
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
CursorThe AI Code Editor
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.
Amap Maps高德地图官方 MCP Server
RedisA Model Context Protocol server that provides access to Redis databases. This server enables LLMs to interact with Redis key-value stores through a set of standardized tools.
Serper MCP ServerA Serper MCP Server
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
Y GuiA web-based graphical interface for AI chat interactions with support for multiple AI models and MCP (Model Context Protocol) servers.
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
Baidu Map百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。