MakeMeBetter AI MCP Server
Connects AI assistants to your MakeMeBetter AI board, enabling natural-language management of tasks, habits, and goals with read/write tools.
README
MakeMeBetter AI — MCP Server
Connect your AI assistant to your MakeMeBetter AI board and manage your tasks, habits, and goals by talking to it.
Ask Claude "what's on my board today?", "push the quarterly report to Friday at 5pm", or "add buy milk under my groceries task" — and it happens on your real board, in the app on your phone.
This repository holds the public metadata and documentation for the hosted MakeMeBetter AI MCP server. The server itself is a managed service — there is nothing to install or self-host.
| Connector URL | https://makemebetter-ai-backend-59998332584.asia-south1.run.app/mcp |
| Transport | Streamable HTTP |
| Auth | OAuth 2.1 — you sign in and approve access; no API keys to copy around |
| Tools | 11 (4 read, 7 write) |
| Cost | Free tier included with every MakeMeBetter AI account |
Getting started
You need a MakeMeBetter AI account. Sign in to the consent screen with the same email as your app account so your assistant lands on your existing board.
Claude (web, desktop, mobile)
Settings → Connectors → Add custom connector → paste the connector URL → sign in → approve. Added once, it works across every Claude surface.
Claude Code
claude mcp add --transport http makemebetter https://makemebetter-ai-backend-59998332584.asia-south1.run.app/mcp
Then complete the OAuth flow in your browser.
ChatGPT
Requires a paid plan (Plus/Pro/Business/Enterprise) and Developer mode (Settings → Apps & Connectors → Advanced settings → Developer mode). The option to add a custom connector only appears once both are true. Then add the connector URL and sign in.
Cursor / VS Code
Add to your editor's MCP config:
{
"mcpServers": {
"makemebetter": {
"url": "https://makemebetter-ai-backend-59998332584.asia-south1.run.app/mcp"
}
}
}
The editor drives the OAuth flow for you.
Tools
Records come back with id, title, status (not_started / in_progress / completed) and
progress_percent, plus due_date, due_time, target_frequency and notes when they're set.
Reading
| Tool | What it does |
|---|---|
list_tasks(include_completed?) |
Your top-level tasks. Active only by default. |
list_habits(include_completed?) |
Your habits, with target_frequency when set. |
list_goals(include_completed?) |
Your longer-term goals. |
get_record(record_id) |
Full detail for one record. For a task, includes its subtasks. |
Writing
| Tool | What it does |
|---|---|
create_task(title, due_date?, due_time?, notes?, parent_task_id?) |
Creates a task, or a subtask when you pass parent_task_id. |
create_habit(title, target_frequency?, notes?) |
target_frequency is free text — "daily", "3x per week". |
create_goal(title, notes?) |
Creates a goal. |
update_record(record_id, ...) |
Partial update — only the fields you pass change. |
set_progress(record_id, percentage) |
Sets progress 0–100. |
complete_record(record_id) |
Marks completed at 100%. |
delete_record(record_id) |
Permanent, and cascades to a task's subtasks. Assistants are instructed to confirm with you first. |
Dates are ISO (2026-07-15); times are 24-hour (17:30) and require a date.
Privacy and security
- Your board only. Every read and write is filtered by your authenticated user id at the database query level. Another user's record id behaves exactly like one that doesn't exist — the server never reveals whether it exists.
- No credentials to hand over. Authentication is OAuth 2.1; your assistant receives a scoped access token, never your password.
- Discovery follows RFC 9728 at
/.well-known/oauth-protected-resource/mcp.
Usage limits
Tool calls are metered per day, per account:
| Plan | Tool calls / day |
|---|---|
| Free | 25 |
| Pro | 2,000 |
initialize and tools/list are free — only actual tool calls count. When you run out, tools
return a plain-language limit notice instead of data, and reset the next UTC day.
Support
Issues and questions: open an issue.
License
The metadata and documentation in this repository are released under the MIT License. The MakeMeBetter AI service itself is a separate, proprietary hosted product.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.