Connect Anteriad to ChatGPT: Research Intent and Account Matches

You want to connect Anteriad to ChatGPT so your revenue operations teams, sales engineers, and marketing AI agents can research B2B buyer intent, match account records, and extract contact volumes using natural language. If your team uses Claude instead, check out our guide on connecting Anteriad to Claude, or explore our broader architectural overview on connecting Anteriad to AI Agents.

Giving a Large Language Model (LLM) read access to a sprawling B2B identity graph like Anteriad is a serious engineering challenge. You are dealing with complex account linkages, intent topics, scoring thresholds, and strict API constraints. You either spend weeks building, hosting, and maintaining a custom Model Context Protocol (MCP) server, or you use a managed infrastructure layer that handles the boilerplate for you.

This guide breaks down exactly how to use Truto to generate a secure, managed MCP server for Anteriad, connect it natively to ChatGPT, and execute complex account research workflows using conversational prompts.

The Engineering Reality of the Anteriad API

A custom MCP server is a self-hosted integration layer that translates an LLM's tool calls into REST API requests. While the open MCP standard provides a predictable way for models to discover tools over JSON-RPC 2.0, implementing it against vendor APIs is painful. If you decide to build a custom MCP server for Anteriad, you own the entire API lifecycle. Here are the specific integration challenges that break standard CRUD assumptions when working with Anteriad:

The B2B Identity Graph Maze

Anteriad does not return a simple, flat list of "companies." It relies on an intricate identity graph composed of Account Links, Intent Topics, and Xplorer records. To get a complete picture of an account, an LLM cannot just query a single endpoint. It must first resolve a company name and address to an Anteriad Match ID, then look up the associated Account Link ID, and finally query the contact counts or intent scores associated with that link. If your custom server does not expose these endpoints with explicitly mapped JSON schemas, the LLM will fail to understand the required sequence and hallucinate the relational lookups.

Rate Limits and the 429 Reality

When an LLM attempts to analyze intent across 50 different domains, it will fire off rapid, concurrent requests. Anteriad enforces strict rate limits to protect its infrastructure. When you hit these limits, the upstream API returns an HTTP 429 Too Many Requests error.

It is critical to understand how this is handled: Truto does not automatically retry, throttle, or apply backoff on rate limit errors. Instead, Truto passes that 429 error directly back to the caller (your MCP client) while normalizing the upstream rate limit information into standardized IETF headers (ratelimit-limit, ratelimit-remaining, ratelimit-reset). The caller - or the agent framework wrapping the LLM - is entirely responsible for reading these headers and implementing the appropriate exponential backoff or retry logic.

Pagination and Cursor Enforcement

When an LLM requests a list of intent topics or account links, it cannot ingest 100,000 records at once. You must write the logic to handle pagination cursors. The MCP server must inject explicit instructions into the tool schema, telling the LLM to pass the next_cursor values back unchanged to fetch the subsequent set of records. If the schema is ambiguous, the LLM will invent pagination parameters or summarize truncated data as if it were the complete dataset.

How to Generate the Anteriad MCP Server

Instead of writing and hosting the JSON-RPC translation layer yourself, you can use Truto to dynamically generate a secure MCP server URL. Truto derives the tool definitions directly from the integration's resource schemas and documentation records, ensuring the LLM always has accurate parameters.

You can generate the MCP server using either the Truto UI or the API.

Method 1: Via the Truto UI

The fastest way to get started is to use the Truto dashboard to generate a server for a specific integrated account.

1. Log in to your Truto environment and navigate to the integrated account page for your Anteriad connection.
2. Click the MCP Servers tab.
3. Click Create MCP Server.
4. Select your desired configuration (e.g., restrict methods to read-only, assign specific tags, or set an expiration date).
5. Click Save and copy the generated MCP server URL. You will need this URL to connect ChatGPT.

Method 2: Via the Truto API

For platform engineering teams building programmatic agent deployments, you can provision MCP servers dynamically via the Truto API. This creates a dedicated token and URL for the specified integrated account.

Make an authenticated POST request to the /integrated-account/:id/mcp endpoint:

curl -X POST "https://api.truto.one/integrated-account/YOUR_ACCOUNT_ID/mcp" \
  -H "Authorization: Bearer YOUR_TRUTO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Anteriad RevOps AI Server",
    "config": {
      "methods": ["read", "list", "get"]
    }
  }'

The API will return a JSON payload containing the secure server URL:

{
  "id": "mcp-token-xyz-789",
  "name": "Anteriad RevOps AI Server",
  "config": { "methods": ["read", "list", "get"] },
  "expires_at": null,
  "url": "https://api.truto.one/mcp/a1b2c3d4e5f6g7h8..."
}

How to Connect the MCP Server to ChatGPT

Once you have your Truto MCP server URL, you must connect it to your ChatGPT instance. Truto's servers are fully self-contained; the token in the URL handles the routing and authentication for the specific Anteriad account. You can connect via the ChatGPT interface or via a local configuration file for programmatic testing.

Method A: Via the ChatGPT UI

If you are using ChatGPT Enterprise, Pro, Plus, Business, or Education, you can add custom connectors directly through the settings interface.

1. In ChatGPT, click your profile picture and navigate to Settings -> Apps -> Advanced settings.
2. Toggle Developer mode to the ON position (MCP functionality is currently behind this flag).
3. Under the MCP servers / Custom connectors section, click Add a new server.
4. Enter a descriptive Name (e.g., "Anteriad B2B Data").
5. In the Server URL field, paste the full URL you generated from Truto (https://api.truto.one/mcp/...).
6. Click Save.

ChatGPT will immediately handshake with the Truto server, securely fetch the available Anteriad tools, and expose them to your conversational interface.

Method B: Via Manual Config File

If you are testing locally or integrating ChatGPT via a programmatic multi-agent framework that uses standard MCP configurations, you can define the server using an SSE (Server-Sent Events) transport file.

Create a file named chatgpt_mcp_config.json and add the following configuration:

{
  "mcpServers": {
    "anteriad-data-server": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sse",
        "--url",
        "https://api.truto.one/mcp/YOUR_SECURE_TOKEN"
      ]
    }
  }
}

This configuration instructs your local MCP client to connect to the Truto server over HTTP using the SSE transport protocol, seamlessly exposing the Anteriad tools to your environment.

Anteriad Hero Tools for ChatGPT

Truto automatically translates Anteriad's API resources into discrete, documented tools for the LLM. Rather than exposing every single granular endpoint, here are the highest-leverage operations your AI agents can perform.

1. List All Anteriad Intent

Tool name: list_all_anteriad_intent

This tool allows ChatGPT to query Anteriad for companies showing active buying signals for specific topics. It returns the best-matched companies based on intent signals, including the company ID and detailed intent attributes. You must provide the topic parameter.

"Find companies showing high intent for the topic 'Cloud Security Migration' and list their Anteriad IDs."

2. List All Anteriad Account Links

Tool name: list_all_anteriad_account_link

This tool is critical for resolving network data back to a concrete business entity. It searches for account links based on criteria like cidr, ip, domain, score, or threshold. This allows the LLM to identify which corporate account owns a specific IP block or website.

"We saw high traffic from the domain 'acmecorp.com'. Look up the Anteriad account link for this domain and return the account link ID."

3. List All Anteriad Match

Tool name: list_all_anteriad_match

Identity resolution is the hardest part of B2B data. This tool allows the LLM to take a raw company name and physical address and resolve it to a canonical Anteriad matched company record. It returns the matched company ID and specific firmographic attributes.

"Match the company 'Global Tech Industries' located at '123 Innovation Drive, San Francisco, CA' in Anteriad and return their exact company record."

4. List All Anteriad Intent Topics

Tool name: list_all_anteriad_intent_topics

Unlike the standard intent tool, this operation returns the best matched company specifically formatted for a given intent topic, including the company name, address, topic string, and the critical accountLinkId. You can optionally apply threshold and score filters to eliminate low-intent noise.

"Search for the intent topic 'Enterprise Resource Planning' and show me the top matched companies with an intent score above 85."

5. List All Anteriad Xplorer Contact Counts

Tool name: list_all_anteriad_xplorer_contact_counts

Knowing an account has intent is useless if you do not know if they have enough personnel to target. This tool queries Anteriad's Xplorer database to return the total available contact counts for a specific account. It requires an account_link_id.

"Take the account link ID 'AL-98765' and check the Anteriad Xplorer contact counts so we know how many potential leads exist in that account."

For the complete inventory of available tools, including MAID counts and cookie aggregations, view the Anteriad integration page.

Workflows in Action

Individual tools are useful, but the real power of an MCP server is allowing the LLM to string multiple API calls together to accomplish complex research workflows. Here are two real-world scenarios showing how ChatGPT orchestrates Anteriad tools.

Workflow 1: Intent-Driven Target Account Discovery

Persona: Sales Development Representative (SDR)

An SDR wants to build a highly targeted outreach list based on immediate buying signals, rather than just guessing who to email.

"Find the top companies showing intent for the topic 'Zero Trust Architecture'. Once you have their names, resolve their domains to Account Link IDs, and then check how many total contacts we can source for those specific accounts."

Execution Steps:

1. list_all_anteriad_intent_topics: ChatGPT queries the topic "Zero Trust Architecture" to retrieve the top matched companies and their associated accountLinkIds.
2. list_all_anteriad_xplorer_contact_counts: The LLM iterates through the retrieved accountLinkIds, querying Anteriad to extract the exact number of available contacts for each company.

Result: The SDR receives a prioritized list of companies actively researching Zero Trust, alongside the exact number of contacts available in the Anteriad database, allowing them to prioritize accounts with the highest reach.

sequenceDiagram
    participant User as Sales Rep
    participant ChatGPT as ChatGPT
    participant Truto as Truto MCP Server
    participant Anteriad as Anteriad API

    User->>ChatGPT: "Find Zero Trust intent, get account links, check contact counts."
    ChatGPT->>Truto: Call list_all_anteriad_intent_topics(topic: "Zero Trust")
    Truto->>Anteriad: GET /intent/topics
    Anteriad-->>Truto: Returns matched companies & AccountLinkIDs
    Truto-->>ChatGPT: JSON result (AccountLinkIDs)
    
    loop For each AccountLinkID
        ChatGPT->>Truto: Call list_all_anteriad_xplorer_contact_counts(account_link_id)
        Truto->>Anteriad: GET /xplorer/contacts/counts
        Anteriad-->>Truto: Returns contact volume
        Truto-->>ChatGPT: JSON result (Contact counts)
    end
    
    ChatGPT-->>User: Final report with intent accounts and contact volumes.

Workflow 2: Account Enrichment and Link Resolution

Persona: Revenue Operations (RevOps)

A RevOps manager has a messy list of physical addresses and company names from a recent trade show and needs to map them to canonical Anteriad IDs to check if they are showing intent for the company's product.

"I have a list of physical addresses and company names from our last event. Match them in Anteriad to get their canonical IDs, and then tell me if those specific companies are showing intent for 'Data Warehousing'."

Execution Steps:

1. list_all_anteriad_match: ChatGPT passes the raw company names and addresses to the matching engine, resolving the messy data into canonical Anteriad company IDs and Account Links.
2. list_all_anteriad_intent: The LLM then queries the intent endpoints for those specific Account Links to see if they index highly for the "Data Warehousing" topic.

Result: The messy spreadsheet data is normalized into strict Anteriad entities, enriched with current intent signals, allowing the marketing team to immediately trigger targeted ad campaigns.

flowchart LR
    A["Raw Event Data<br>(Name & Address)"] --> B["ChatGPT Agent<br>(RevOps)"]
    B -->|"list_all_anteriad_match"| C["Canonical Anteriad ID<br>& Account Link"]
    C -->|"list_all_anteriad_intent"| D["Intent Signal Check<br>('Data Warehousing')"]
    D --> E["Enriched Target List<br>Ready for Marketing"]

Security and Access Control

Exposing B2B data repositories to an LLM requires strict governance. You do not want a general-purpose AI agent accidentally triggering massive data scraping jobs. Truto MCP servers provide multiple layers of configuration to lock down access:

• Method Filtering: When creating the server via UI or API, you can restrict the token to specific operations. By passing "methods": ["read"] in the configuration, you ensure the LLM can only execute get and list operations, preventing any accidental write or delete commands.
• Tag Filtering: You can restrict the MCP server to only expose tools associated with specific functional areas (e.g., "tags": ["intent"]), hiding unrelated endpoints like cookie metrics or MAID counts from the LLM.
• API Token Authentication: By setting require_api_token_auth: true, possession of the MCP URL is no longer sufficient to access the tools. The client must also pass a valid Truto API token in the Authorization header, enforcing a dual-layer security model.
• Ephemeral Servers: You can pass an ISO datetime to the expires_at field when creating the server. Truto enforces this via Cloudflare KV expiration and a scheduled Durable Object cleanup alarm, automatically permanently destroying the server and its tokens once the TTL expires. This is ideal for short-lived research agents.

Moving from Manual Research to Agentic Ops

B2B revenue teams spend thousands of hours manually cross-referencing CSV files against identity graphs to find the right accounts to target. By connecting Anteriad to ChatGPT via a managed MCP server, you eliminate the manual swivel-chair operations.

Instead of dealing with rate limit backoffs, nested payloads, and pagination cursors, your engineers can let Truto handle the JSON-RPC translation. This allows your Go-To-Market teams to simply converse with their data, requesting intent signals and contact volumes in plain English, and executing complex account research in seconds.