Skip to main content
Resources
Agent Grounding Handbooks

AI Agent Grounding: Connecting to the GraphIQ.ai MCP Server.

The Model Context Protocol (MCP) is an open standard that lets AI assistants and automated agents call our platform data tools natively. This guide outlines how to plug our HTTP streamable server (app.graphiq.ai/mcp) directly into active LLM reasoning loops. Grounding your models in a pre-resolved graph of 287M organizations gives them resolved entities to reason over instead of scraped text. With MCP 2.0, agents get 17 read-only tools on connection across organizations, people, news, ontologies, web and guides, plus 6 built-in playbooks. MCP sign-in uses OAuth with your GraphIQ.ai account. Read the MCP 2.0 release notes →

What your agent gets

17 read-only tools in six groups.

Your agent discovers every tool when it connects and picks the right ones for your prompt, so you don’t need to know the tool names. All 17 tools are read-only, so an agent can explore freely without changing your data.

OrganizationsSearch, full company profiles and market counts.
PeopleContacts filtered by seniority and function.
NewsDeduplicated coverage of a company, with sentiment scores.
OntologiesCapabilities, technologies, NAICS codes and places.
WebLive web search and page reading, returned clean and compact.
GuidesStep-by-step playbooks for complex tasks.
6 built-in playbooks:Company researchMarket mappingProspect listsFind contactsNews monitoringQuery syntaxHow playbooks work →
Configuration

Three ways to connect.

PATH 01

Chat apps and agents

Claude, ChatGPT, Microsoft Copilot Studio and any AI tool that supports remote MCP servers. Add GraphIQ.ai as a remote MCP server in the tool’s connector or MCP settings, then sign in with your GraphIQ.ai account. A paid MCP account is required.

ClaudeChatGPTCopilot Studio
connector settings
Server type  Remote MCP server
URL          https://app.graphiq.ai/mcp
Sign-in      OAuth → your GraphIQ.ai account

# then ask in plain English, for example:
# “Use the prospect lists guide to build me
#  a territory list of…”
PATH 02

Coding tools

Claude Code, Cursor and other MCP clients that read a config file. Add the GraphIQ.ai server URL to the client’s MCP config. No key goes in the file: the client opens a browser window to sign in with OAuth the first time it connects.

Claude CodeCursor
.mcp.json
{ "mcpServers": {
    "graphiq": {
      "type": "http",
      "url": "https://app.graphiq.ai/mcp"
    }
} }

# no key in the file: sign in with OAuth on first use
PATH 03

Systematic access with the REST API

The REST API is the right fit when your code decides what to fetch and when: CRM enrichment jobs, data pipelines, scheduled queries and coding agents such as Claude Code and Codex. It uses Bearer authentication. See the API quickstart for endpoints and samples.

REST APIBearer auth
request headers
Authorization: Bearer $GRAPHIQ_KEY
Content-Type: application/json

# MCP  → OAuth, for chat-based agents
# REST → Bearer, for code and pipelines
Security

Account identifier & data security.

MCP and the REST API use different sign-in methods. Every MCP session runs under an authorized OAuth token, and all 17 MCP tools are read-only.

REST API · Bearer token

Scripts, background jobs, server-to-server integrations and coding tools call the REST API with a Bearer token in the Authorization header.

MCP · OAuth 2.0

Chat apps, agents and coding tools connect to the MCP server through a browser OAuth 2.0 sign-in with your GraphIQ.ai account.

Connect the dots

Where MCP agent access connects your GTM.

Talk to our team

Connect your AI agents to a source of data drawn from public sources, with a confidence score.

Talk to our team to initialize your MCP server connection, run a test query against 287M resolved organizations, and verify your agent toolchain configuration end-to-end.

Book a demo