How it works
- Connect AB Tasty's MCP server to your AI assistant or coding agent following the @abtasty/mcp-server setup, or paste a web campaign plan into the chat.
- 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 collects the plan from the MCP server, the API, or your paste, names the path it used, then shapes the text into hypothesis, versions, audience, traffic split, main goal, and guardrails.
- It sends that text to ExpLint with
explint__lint, which returns errors, warnings, and passes, each finding naming the rule behind it. - You update the test in AB Tasty, bring the plan back the same way, and ask for a second lint that ties any remaining finding to the setup step behind it.
Copy and paste this prompt
Change the words in square brackets to name your own project, board, or test.
Prompt for your assistant
Here is a test plan from AB Tasty, from its MCP server, its API, or pasted below. Repeat it back headed hypothesis, versions and what each changes, audience, traffic split, main goal, guardrail goals, and run time, and say which headings are empty. Lint it with explint.com against [your ExpLint project name]. Show each error and warning and the words that caused it. Do not launch anything or invent numbers. Suggest fixes and lint again on my updated plan. Plan: [paste or name the AB Tasty test]
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/names-metric-and-direction | error | The test has to name its main goal and the direction it expects, since AB Tasty records the goal as a setting and the plan is where the expectation goes. |
| variants/control-is-described | error | The plan should say what visitors see today before it describes the variation, or a reader cannot tell what the test is measuring against. |
| randomization/states-the-split | error | A traffic split set in the campaign still belongs in the plan text, because the reader judging the run length needs to see the share. |
| decision/has-criteria-before-launch | error | The plan should record what result would ship the variation and what result would send it back, before the campaign collects its first visitor. |
