Skip to main content

Getting Support

This guide provides information on how to get help with the Model Context Protocol.

Support Channels​

Discord Community​

Our Discord server is the best place to get real-time help from the community and the TRMX team:

  • Join: TRMX Discord Server
  • Channels:
    • #general: General MCP discussion
    • #help: Get help with MCP issues
    • #announcements: Important announcements
    • #showcase: Share what you've built

GitHub Issues​

For bug reports and feature requests, use our GitHub repositories:

Community Forum​

Our community forum is great for longer discussions and knowledge sharing:

  • Forum: community.trmx.ai
  • Categories:
    • Technical Questions
    • Showcase
    • Feature Requests
    • General Discussion

Stack Overflow​

For technical questions, you can also use Stack Overflow with the trmx-mcp tag:

How to Ask for Help​

When asking for help, please include:

  1. MCP Version: Client/Server SDK version you're using
  2. Environment: Node.js version, OS, etc.
  3. Problem Description: Clear description of the issue
  4. Code Sample: Minimal reproducible example
  5. Error Messages: Full error messages and stack traces
  6. Steps Tried: What you've already attempted

Example Support Request​

Title: Error when sending messages with attachments

MCP Client SDK Version: 1.2.0
Node.js Version: 18.12.1
OS: macOS 12.6

I'm trying to send a message with an image attachment, but I'm getting the following error:

"Error: Attachment data must be base64 encoded"

Code:

const response = await conversation.sendMessage({
content: "What is in this image?",
attachments: [
{
type: "image",
mimeType: "image/jpeg",
data: imageData, // This is a Buffer from fs.readFile
},
],
});

Full error:
Error: Attachment data must be base64 encoded
at validateAttachment (/node_modules/@trmx/client/dist/utils/validation.js:45:11)
at validateMessage (/node_modules/@trmx/client/dist/utils/validation.js:78:5)
at Conversation.sendMessage (/node_modules/@trmx/client/dist/conversation.js:102:3)

I've tried providing a URL instead, but I need to use local files.

Support Tiers​

Community Support​

  • Discord Community: Free, peer-to-peer support
  • GitHub Issues: Bug reporting and feature requests
  • Community Forum: Knowledge sharing and discussions
  • Documentation: Self-service help through guides and references

TRMX Cloud Support​

TRMX Cloud users receive additional support based on their plan:

  • Developer Plan: Email support with 48-hour response time
  • Team Plan: Email support with 24-hour response time
  • Enterprise Plan: Dedicated support manager, phone support, and guaranteed SLAs

Training & Education​

Workshops​

We offer regular online workshops on MCP development:

  • Introduction to MCP: Monthly beginner workshop
  • Advanced MCP: Quarterly deep-dive sessions
  • Custom Workshops: Available for Enterprise customers

Office Hours​

Join our bi-weekly office hours with the TRMX engineering team:

FAQs​

How do I reset a conversation context?

Use the resetContext() method on the conversation object:

await conversation.resetContext({
systemPrompt: "New system prompt for reset conversation"
});

How do I enable streaming responses?

Set the streaming option when creating a conversation:

const conversation = client.createConversation({
contextId: "user-123",
streaming: true
});

const stream = await conversation.sendMessage("Generate a story");
stream.on("data", chunk => console.log(chunk.content));

How do I authenticate with OAuth?

Use the auth option when creating a client:

const client = new MCPClient({
serverUrl: "https://your-mcp-server.trmx.ai",
auth: {
type: "oauth",
token: "your-oauth-token"
}
});

Where are conversation contexts stored?

By default, contexts are stored in memory on the MCP server. For production, you should configure a persistent store like MongoDB or Redis.

Contributing to the Community​

You can help improve the MCP community:

  1. Answer Questions: Help others in Discord or the forum
  2. Share Examples: Post code samples and templates
  3. Write Articles: Blog about your MCP experiences
  4. Report Bugs: Submit detailed bug reports
  5. Suggest Improvements: Share ideas for improving MCP

Next Steps​