How it works
- Connect the Notion MCP server at https://mcp.notion.com/mcp to your AI assistant or coding agent, then authorize the connection with OAuth so the assistant reads the pages and databases your Notion account can already reach in the workspace you select.
- Install the ExpLint MCP at https://explint.com/mcp to enable your agent to audit tests.
- Run the prompt below to audit these plans before the work starts.
How the check runs
- It searches the pages and databases you name for briefs whose status property marks them as still being drafted, joins the title, page content, and status into one text, and names any brief that arrives too thin to check.
- It sends that text to ExpLint with
explint__lint, which returns errors, warnings, and passes, each finding naming the rule behind it. - When you approve a brief, it adds the findings and a suggested rewrite as a comment on the page and lints the rewrite, so the brief's status can move once the errors clear.
Copy and paste this prompt
Change the words in square brackets to name your own project, board, or test.
Prompt for your assistant
Use the Notion MCP tools to search [workspace or database name] for experiment briefs whose status property is [Draft]. For each brief, fetch the title, full page content, and database properties, then lint that text with explint.com against project [your ExpLint project name]. Show me a table with the brief title, its errors and warnings, and the exact words behind each finding. Do not change any page yet. When I approve a brief, add the findings and a suggested rewrite as a comment on its page, then lint the rewrite and show whether the errors are gone.
What ExpLint looks for
ExpLint ships over 24 rules, and a project switches each one on or off or rewrites it in plain English. These are the rules the plans this guide reads break most often. An error means the test cannot answer its question as written, and a warning means the result will be hard to judge when it comes in.
| Rule | Level | In plain words |
|---|---|---|
| hypothesis/is-falsifiable | error | A brief that only argues the change will help leaves the page with nothing a finished test could contradict. |
| hypothesis/names-metric-and-direction | error | The brief has to name the number and the direction it expects, or the status property becomes the only thing marking it as a test. |
| hypothesis/explains-the-mechanism | warning | One or two sentences on why the change should move people belong in the brief, since a reader new to the idea otherwise has to guess the reasoning. |
| hypothesis/is-based-on-evidence | error | A brief with no link to research, usage data, or an earlier test reads as an opinion, which the lint marks and the status property then carries into planning. |
