Connect Membes to ChatGPT: Sync Members, Events, and CPD Activities

If you need to connect Membes to ChatGPT to automate association management, member directories, or event registrations, you need a Model Context Protocol (MCP) server. This server acts as the translation layer between an LLM's function calls and the underlying REST architecture of Membes. If your team uses Claude, check out our guide on connecting Membes to Claude or explore our broader architectural overview on connecting Membes to AI Agents.

Giving an AI agent read and write access to an Association Management System (AMS) like Membes is an engineering challenge. You either spend weeks building, hosting, and maintaining a custom MCP server, writing JSON schemas for every endpoint, and handling OAuth token refreshes - or you use a managed infrastructure layer that dynamically generates a secure, authenticated MCP server URL for you.

This guide breaks down exactly how to use Truto to generate a managed MCP server for Membes, connect it natively to ChatGPT, and execute complex membership workflows using natural language.

The Engineering Reality of the Membes API

A custom MCP server is a self-hosted integration layer. While the open MCP standard provides a predictable way for models to discover tools, the reality of implementing it against a domain-specific vendor API like Membes is painful. You aren't just building generic CRUD operations - you are integrating with an AMS that has highly specific business logic.

If you decide to build a custom MCP server for Membes, you own the entire API lifecycle. Here are the specific integration challenges that break standard assumptions when working with Membes:

The CPD Validation Maze Continuing Professional Development (CPD) is core to Membes. Writing a CPD log to a member's profile is not a simple database insert. The Membes API requires strict adherence to category rules. Before you can log a CPD activity, you must validate the category_id against specific point caps (cap_points2, cap_points5) and minimum requirements (min_hours, min_points3). If your custom server doesn't expose the list_all_membes_cpd_activities schema to the LLM first, the LLM will hallucinate category IDs, submit invalid data, and fail silently.

Identity and Profile Fragmentation Membes splits profile data across multiple functional areas. A standard profile lookup (get_single_membes_profile_by_id) returns core demographics, but if an AI agent needs to understand a member's interaction history or batch data, it has to orchestrate calls across membes_profiles_search, list_all_membes_batch_profile_interactions, and directory mapping endpoints. A custom server requires you to write complex aggregation logic so the LLM can make sense of a single member.

Strict 429 Pass-Throughs and Backoff Handling Membes enforces rate limits to protect its infrastructure. It is critical to understand that Truto does not retry, throttle, or apply backoff on rate limit errors. When the Membes upstream API returns an HTTP 429 Too Many Requests, Truto normalizes that upstream rate limit info into standardized headers (ratelimit-limit, ratelimit-remaining, ratelimit-reset) per the IETF spec, and passes the error directly back to ChatGPT. Your LLM prompt instructions must explicitly tell the agent to respect these headers and retry intelligently. If your custom server attempts to infinitely retry, Membes will hard-block the application.

Generating the Membes MCP Server

Instead of writing integration code, you can use Truto to dynamically generate an MCP server mapped directly to a connected Membes account. Truto handles the token lifecycle, normalizes the pagination, and dynamically generates the tool schemas.

There are two ways to generate your Membes MCP server: via the Truto UI or programmatically via the API.

Method 1: Via the Truto UI

For teams testing workflows or setting up internal automations, the UI is the fastest path.

1. Navigate to the Integrated Accounts page in your Truto dashboard.
2. Select your connected Membes instance.
3. Click the MCP Servers tab.
4. Click Create MCP Server.
5. Select your desired configuration (e.g., allowing only read methods, or scoping to specific tags).
6. Copy the generated MCP server URL. It will look like https://api.truto.one/mcp/a1b2c3d4e5f6...

Method 2: Via the API

If you are provisioning AI agents programmatically for your users, you should generate MCP servers via the Truto REST API. This creates a secure, hashed token in KV storage linked specifically to that tenant's Membes instance.

Make a POST request to /integrated-account/:id/mcp:

const response = await fetch('https://api.truto.one/integrated-account/YOUR_MEMBES_ACCOUNT_ID/mcp', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_TRUTO_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    name: "Membes Triage Agent",
    config: {
      methods: ["read", "write"], // Exposes full CRUD capabilities
      require_api_token_auth: false
    },
    expires_at: "2026-12-31T23:59:59Z" // Optional: auto-revoke access
  })
});
 
const data = await response.json();
console.log(data.url); // The MCP server URL to pass to ChatGPT

The returned URL is fully self-contained. The token in the path securely encodes the environment, the specific Membes connection, and the filter configurations.

Connecting the Membes MCP Server to ChatGPT

Once you have your Truto MCP server URL, you must connect it to your LLM framework. ChatGPT natively supports remote MCP servers.

Method A: Via the ChatGPT UI

If you are using ChatGPT Pro, Plus, Enterprise, or Education accounts, you can add the connector directly in the application interface:

1. Open ChatGPT and go to Settings.
2. Navigate to Apps -> Advanced settings.
3. Toggle on Developer mode (MCP support requires this flag).
4. Under MCP servers / Custom connectors, select Add a new server.
5. Set the Name to something recognizable (e.g., "Membes AMS").
6. Paste the Truto MCP URL into the Server URL field.
7. Click Save.

ChatGPT will immediately ping the /initialize endpoint, discover the Membes tool schemas, and make them available in your current conversation.

Method B: Via Manual Config File

If you are running a local MCP client, a headless agentic framework, or standardizing configurations across a dev team, you can use a manual JSON configuration file. You will use the standard @modelcontextprotocol/server-sse package to wrap the remote HTTP connection.

Create or update your MCP configuration file (e.g., mcp-config.json or your framework's equivalent):

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

When your agent boots, it will execute this npx command, establishing a Server-Sent Events (SSE) connection to Truto's router, allowing bidirectional JSON-RPC 2.0 communication with the Membes API.

Membes Hero Tools

Truto derives tools dynamically from the underlying Membes documentation records. Here are the highest-leverage tools available for ChatGPT, completely bypassing the need to write custom integration code.

membes_profiles_search

This is the core discovery tool. Membes allows searching by email, radius from a location, or custom fields.

Contextual note: The Membes API dictates that only one search criteria is evaluated at a time. If the LLM passes an email and a location radius in the same payload, Membes ignores the secondary parameters. Ensure your prompts instruct the LLM to search by one primary attribute.

"Find the Membes profile associated with sarah.connor@example.com. Extract her profile_number and current membership_paid_through date."

get_single_membes_profile_by_id

Once an agent identifies a profile from a search, it uses this tool to retrieve the complete data payload, including custom fields, address groups, and nested member attributes.

Contextual note: This is a heavy payload. If the LLM is doing batch analysis, instruct it to use the list_all_membes_batch_profiles tool instead to prevent context window overflow.

"Look up the full profile details for member ID 98234. Check if they have any active warnings in their custom_fields array."

list_all_membes_group_events

Retrieves the calendar of events scoped to a specific group ID.

Contextual note: This tool is crucial for checking registration_open and member_only flags before an AI agent attempts to register a user for an upcoming seminar.

"Fetch all upcoming events for group ID 12. Filter the response in your output to only show events where registration_open is true."

create_a_membes_event_registration

Registers an existing profile for an event.

Contextual note: You must pass the exact event_id, profile_id, and registration_type. The LLM should typically call membes_events_get_registration_types beforehand to discover the valid registration_type strings for the specific event.

"Register profile ID 98234 for event ID 405. Use the standard member registration type string you found in your previous query."

list_all_membes_cpd_activities

Retrieves the complex matrix of available Continuing Professional Development activities and their validation rules (caps, min hours, point requirements).

Contextual note: AI agents must use this tool as a lookup table to validate CPD point submissions before attempting to push data, preventing API rejections.

"Get the list of all valid CPD activities. Find the category_id for 'Online Webinar' and note the cap_points2 limit."

create_a_membes_cpd_activity_log

Writes a CPD log directly to a member's profile.

Contextual note: Requires name, category_id, date, profile, and activity details. The LLM must meticulously map the data to the schema requirements discovered in the previous step.

"Log a new CPD activity for profile 98234. Use category_id 5, set the date to today, and log 2 points for attending the Annual Compliance Seminar."

For a complete list of tools, including forum management, news publishing, and directory filtering, view the Membes integration page.

Workflows in Action

Once the MCP server is connected, ChatGPT can orchestrate multi-step business logic by chaining tools together. Here are two real-world examples of how AI agents interact with the Membes API.

Scenario 1: Autonomous Event Registration and CPD Allocation

A professional association receives an email from a member requesting to be registered for a specific seminar, and asking for their CPD points to be updated automatically.

"Sarah Connor emailed us asking to register for the 'Q3 Ethics Seminar'. Please find her profile, register her for the event, and log 3 CPD points under the Ethics category for her attendance."

Tool Execution Sequence:

1. membes_profiles_search: The LLM searches by Sarah's email to retrieve her profile_id.
2. list_all_membes_group_events: The LLM scans the events calendar to locate the event_id for the "Q3 Ethics Seminar".
3. membes_events_get_registration_types: The LLM queries the specific event to find the correct registration_type ID.
4. create_a_membes_event_registration: The LLM submits the POST request to book her ticket using the gathered IDs.
5. list_all_membes_cpd_activities: The LLM searches the CPD matrix to find the exact category_id for "Ethics".
6. create_a_membes_cpd_activity_log: The LLM successfully writes the 3 points to her profile.

Result: ChatGPT confirms the ticket is booked and the CPD points are logged, completely bypassing the manual admin interface.

sequenceDiagram
    participant LLM as ChatGPT
    participant MCP as Truto MCP Server
    participant AMS as Membes API

    LLM->>MCP: Call tool: membes_profiles_search<br>{"email": "sarah@example.com"}
    MCP->>AMS: GET /api/v1/profiles/search
    AMS-->>MCP: 200 OK (Profile ID: 98234)
    MCP-->>LLM: JSON Profile Data
    
    LLM->>MCP: Call tool: create_a_membes_event_registration<br>{"profile_id": 98234, ...}
    MCP->>AMS: POST /api/v1/events/register
    AMS-->>MCP: 201 Created
    MCP-->>LLM: Success Confirmation

Scenario 2: Member Interaction Triage

An association manager wants to understand a member's recent engagement before reaching out for a renewal conversation.

"Look up the profile for David Smith. Tell me what his membership status is, and summarize all of his profile interactions from the last 6 months."

Tool Execution Sequence:

1. membes_profiles_search: The LLM locates David by name or email.
2. get_single_membes_profile_by_id: The LLM pulls the core profile to check membership_paid_through and membership_status.
3. list_all_membes_profile_interaction_types: The LLM discovers what interaction types exist in this tenant's system.
4. get_single_membes_profile_interaction_by_id: The LLM pulls the interaction history and filters the dates contextually.

Result: ChatGPT returns a clean, human-readable summary of David's recent support tickets, forum posts, and event attendances, giving the manager immediate context for the renewal call.

Security and Access Control

Exposing an AMS to an LLM requires strict governance. Truto's MCP architecture provides native security controls that evaluate on every tool invocation:

• Method Filtering: By configuring the server with config: { methods: ["read"] }, you completely strip all create, update, and delete tools from the LLM's capability list. The model cannot hallucinate a destructive action because the endpoint simply doesn't exist in its context.
• Tag Filtering: You can isolate agent access to specific domains. For example, applying tags: ["events"] ensures the agent can manage registrations but cannot query sensitive financial data or profile custom fields.
• Time-to-Live Expiration: Passing expires_at during server creation stores an absolute Unix timestamp in the underlying KV storage. Once the timestamp passes, the token is automatically purged, instantly severing the LLM's connection to Membes.
• API Token Authorization: By setting require_api_token_auth: true, the MCP server URL alone is useless. The connecting client must also pass a valid Truto API bearer token in the headers, ensuring URL leaks don't compromise your data.

Ship AI Workflows, Not Integration Code

Connecting Membes to ChatGPT manually requires engineering teams to build pagination loops, map complex CPD schemas, and handle fragile token lifecycles. It turns an AI initiative into a legacy integration project.

By utilizing Truto's managed MCP architecture, you shift the burden of maintaining the integration layer to infrastructure. You get dynamic, auto-updating schemas, native security filtering, and standard error handling out of the box, allowing your team to focus entirely on prompting and workflow design.