Skip to main content

Install ChMeetings Resources MCP

MCP URL: https://mcp.chmeetings.com/resources
This endpoint provides resources (API docs), not tools.

 

VS Code & Antigravity

  • Open VS Code / Antigravity
  • Using Windows Press: Ctrl + Shift + P || For Mac Press: Cmd + Shift + P
  • In the command palette, type: "MCP: Add Resource"
  • When prompted for the resource URL, paste: https://mcp.chmeetings.com/resources
  • Press Enter
  • type the MCP name you want like: chmeetings-resources
  • Press Enter to confirm
  • Wait for VS Code to fetch and register the MCP resources.
{
  "inputs": [],
  "servers": {
    "chmeetings-resources": {
      "type": "http",
      "url": "https://mcp.chmeetings.com/resources" 
    }
  }
}

 

Recommendations for Using ChMeetings MCP Resources in IDEs

ChMeetings MCP resources are designed to help developers work faster with AI inside IDEs such as VS Code and other MCP-compatible tools. They can be very effective for exploring APIs, generating modules, editing integrations, and speeding up development workflows.

To get the best results, please keep the following recommendations in mind.

Who should use these resources

ChMeetings MCP resources are most useful for users who are:

  • software engineers and developers
  • technical product builders and integrators
  • advanced users who understand APIs and backend/frontend workflows
  • developers using AI-assisted coding tools inside IDEs

AI can help generate code, explain APIs, refactor modules, and suggest implementations, but it should not be treated as a replacement for engineering judgment. You should be able to review, test, maintain, and continue development after the AI has produced an initial result.

A good rule is:

Use AI to accelerate development, not to replace ownership of the code.

If you are not comfortable reading, debugging, and maintaining code yourself, AI-generated output may create confusion or technical debt.

Recommended development approach

When using ChMeetings MCP resources in your IDE, it is best to work with:

  • a framework you already know well
  • a programming language your team already uses
  • patterns that match your current application architecture

For example, if your team already works with Laravel, .NET, Node.js, NestJS, React, or Vue, keep the AI output aligned with that stack. Avoid asking the AI to generate code in unfamiliar frameworks unless you are prepared to validate and maintain the result.

This usually leads to:

  • cleaner code
  • fewer integration mistakes
  • easier maintenance
  • faster debugging

What MCP resources are for

ChMeetings MCP resources give the AI direct access to focused technical documentation resources. Instead of asking the model to guess how the API works, you can guide it to use the relevant ChMeetings resource.

Examples of available resources may include:

  • people-api-docs
  • groups-api-docs
  • contributions-api-docs
  • and other ChMeetings API resources

Each resource is typically focused on one area of the platform. This helps the AI stay grounded in the correct documentation and produce more accurate outputs.

Best way to use multiple resources

When your task involves more than one domain, use the relevant resources together.

Examples:

  • If you are syncing member data, start with people-api-docs
  • If you are building workflows around groups or segments, include groups-api-docs
  • If you are handling donation or giving features, include contributions-api-docs

This is better than using one broad prompt with no resource direction.

A strong pattern is:

  • identify the feature you want to build
  • identify which ChMeetings API areas are involved
  • point the AI to the matching resource or resources
  • ask for output that fits your actual stack and architecture

How to prompt effectively

The quality of the result depends heavily on the quality of the prompt.

Good prompts are:

  • specific
  • scoped
  • technical
  • grounded in your real app structure
  • clear about language, framework, and desired output

Good prompt structure

A useful prompt usually includes:

1. Context

Explain what you are building.

Example:

I am building a Node.js service that syncs ChMeetings people into our local CRM.

2. Resource to use

Tell the AI which MCP resource is relevant.

Example:

Use the people-api-docs resource.

3. Task

Be explicit about what you want.

Example:

Generate a service module that fetches people records, maps them to our internal model, and handles pagination.

4. Constraints

Tell the AI how the code should look.

Example:

Use TypeScript, keep it framework-agnostic, separate API client logic from mapping logic, and add retry/error handling.

5. Output format

Ask for a practical result.

Example:

Return production-oriented code with a short explanation of the design and note any assumptions.

Example prompts

Here are some strong examples.

Example 1: Generate a module

Use the people-api-docs resource. Generate a TypeScript module for a Node.js backend that fetches people from the ChMeetings API, supports pagination, and maps the response into our internal MemberProfile model. Keep the API client separate from the mapper and add basic error handling.

Example 2: Edit an existing module

Use the groups-api-docs resource. Refactor this service so it can retrieve groups from ChMeetings and cache them for 10 minutes. Keep the existing coding style and do not change the public method names.

Example 3: Build a feature using more than one resource

Use both people-api-docs and groups-api-docs. Create a service that loads a person’s profile and their related groups, then returns a combined DTO for our admin dashboard. Use C# and follow clean architecture principles.

Example 4: Explain before generating

Use the contributions-api-docs resource. First explain the available endpoints relevant to retrieving contribution history, then suggest the best endpoint for generating donor summaries, and finally generate a Laravel service class for that flow.