MCP server · Developer tools

TNL MCP server

Lets your coding AI write, check and follow a plain English plan for each feature before it writes code.

Developer tools Coding For it and development Developer Tools

TNL stands for Typed Natural Language. It is a way to write a short English plan for each feature, approve it once, and have your coding AI stick to it. This MCP server is the part that lets your AI look up those plans, suggest changes to them, and check the code against them.

What is an MCP server? The 30-second version

On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to something, and here it connects your AI to your project's TNL plan files. Once it is connected, your AI can ask the helper to fetch a plan, suggest an edit, or check that the code still matches what you approved. You do not run anything by hand each time. You just ask your AI, and it uses this helper behind the scenes.

What this MCP server does

You ask your AI to build or change a feature. The AI uses this helper to look through your tnl folder and find any plan whose file list overlaps with your request. It shows you a draft plan in the chat and waits for your yes. After you approve, the helper saves the plan to disk and the AI writes code against it. Later, the helper can check that the files and tests named in the plan still exist, and tell you what plans are affected by a change.

Flow diagram: you ask your AI “Draft a plan for adding CSV export to reports”, on your own computer the TNL MCP server works with your computer, and you get back A plan you can read and approve.

Click to zoom

What you can do with it

Find which saved plans are touched by a set of files
Read back the full text of one or more plans by name
Draft a new plan or an edit to an existing one for your review
Save an approved plan to your project
Check that the files and tests named in a plan still exist
Keep a short log of what the AI did during a session

Try asking your AI

“Which TNL plans cover the files I am about to change?”
“Show me the plan for user-rate-limiter”
“Draft a TNL for adding CSV export to the reports page”
“Run the verifier on the tnl folder and tell me what fails”

What it gives back to you

You get back plain text in the chat: a list of plan names, the full text of a plan, a draft plan for you to read, or a short report of what passed and what failed. When a plan is saved, the helper writes a file into your tnl folder. Nothing is written until you approve it.

Before you start

What you need

Node.js 20 or newer
A project folder where you want the tnl directory to live
A coding agent that reads a Markdown instruction file, such as Claude Code, Codex, Gemini or Cursor

Good to know

The helper can write new plan files into your project once you approve a draft, so read the draft before saying yes.

Install it with your AI

Add TNL MCP server to your AI, no technical skills needed

You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.

Sign in to get the install prompt

Members get a ready-made prompt that lets the Claude desktop app check TNL MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.