Skip to main content
News

New in monday apps MCP: instant AI answers from the developer docs

Related products:APIApps FrameworkAI
  • March 10, 2026
  • 1 reply
  • 270 views

alexpo
Forum|alt.badge.img

We just shipped something that changes how you build with the monday apps framework - and it's one of those small additions that quickly becomes indispensable.

When you're building in Cursor, Claude Code, or any MCP-compatible AI assistant, your AI can now search the official monday developer documentation directly - and return structured, source-linked answers without you ever leaving your flow. No more tab-switching mid-build to look up OAuth scopes, GraphQL mutations, API parameters, or SDK patterns.

 

 

What's new

The monday apps MCP connects AI coding assistants directly to the monday apps framework - letting you create apps, manage features, handle deployments, and more through natural language in your IDE.

get_development_context is a new tool in that same MCP that uses Kapa AI's retrieval-augmented generation (RAG) to answer questions from the official monday developer documentation - finding relevant content and generating accurate, source-linked answers.

Questions can come from you directly, or from your AI assistant as part of its dev flow - for example, Cursor agent asking about the right OAuth scopes before generating code.

Rate limits: The tool currently supports 10 queries per 10-minute window. If you hit a rate limit, wait a few minutes before continuing.

 

Example questions

  • "How do I create a board view app feature?"
  • "What OAuth scopes do I need to read and write board items?"
  • "How do I use sessionToken vs OAuth access token for full-stack apps?"
  • "What's the JSON format for updating a status column?"
  • "What parameters does the change_column_value mutation accept?"
  • "What are custom objects and how do I define them?"
  • "How to deploy to monday code?"
  • "Which monday SDK should I use?"

 

How it works in practice

get_development_context works alongside the Apps MCP's action tools - letting your AI assistant combine knowledge with execution. For example, when you prompt: "Create a board view app called Customer Tracker that reads and updates item names" - your AI assistant can look up the required OAuth scopes (boards:read, boards:write) via get_development_context, then use create_app and create_app_feature to scaffold it with the correct permissions.

 

Why it's useful

  • Grounded in official documentation - responses are generated from the official docs and stay close to documented patterns
  • Most responses include working code examples - GraphQL mutations, SDK usage, JavaScript patterns
  • Honest about gaps - when the docs don't cover something, it says so instead of guessing
  • Catches wrong assumptions - if your question contains an incorrect pattern, it redirects to the right approach
  • Flags deprecated APIs and suggests current alternatives

 

What it covers

Official monday.com apps documentation: OAuth scopes, GraphQL API, SDK reference, monday-code deployment, Vibe Design System (high-level), workflow blocks, custom objects, and app development patterns. If something's documented, you can ask about it.

Not designed for: undocumented API behaviors, debugging your specific code, or anything not in the official developer docs.

 

How to get started

Already using the monday apps MCP? The tool is live now - your AI assistant will use it automatically.

Not set up yet? See the documentation. Works with Cursor, Claude Code, VS Code Copilot, Gemini CLI, and any MCP-compatible client.

{
"mcpServers": {
"monday-apps-mcp": {
"command": "npx",
"args": [
"@mondaydotcomorg/monday-api-mcp",
"-t",
"YOUR_API_TOKEN",
"--mode",
"apps"
]
}
}
}

Give it a try with whatever you're working on right now. Drop a reply with questions it handled well, gaps you found, or anything that surprised you - or share feedback via this form.

Your feedback directly shapes how we expand coverage and improve the tool.

1 reply

  • Participating Frequently
  • September 16, 2026
 

The new monday apps MCP update is interesting because get_development_context brings official developer documentation directly into AI-assisted development, so developers can get source-linked answers while building instead of constantly switching between tools.

A similar idea applies to CRM workflows too: connecting AI with business systems can make information and actions available within the workflow where teams already work. For example, AI voice agents can be connected with CRM platforms to handle customer conversations and work with CRM data as part of the same process.

A relevant example is the Zoho CRM AI Voice Agent integration