Lint.ai

Lint.ai is the new, simple way to ensure that every single line of code lives up to the best standards in your codebase. With lint.ai, you can define rules that will run on every commit and every PR, and you can write them using the languages you already know. These rules can quickly and easily:

• Reformat code and push the result back onto the branch:

  
  black . && /L/verbs/commit "chore(reformatting): Apply Black"
  

• Test for common errors, using existing linting tools (any errors or warnings will cause the PR to be rejected):

  
  eslint --max-warnings 1 .
  

• Use arbitrarily complex logic on PRs, like auto-approving a PR if it's just expanding your tests:

  #!/bin/bash
  
  if ! git diff --name-only origin/master | grep -qvE '^test/'; then
    if [ $(git diff --numstat origin/master | awk '{print $1}') -ge \
         $(git diff --numstat origin/master | awk '{print $2}') ]; then
      /L/verbs/stamp "auto-approval by Lint.ai: adding tests"
    else
      # Anything sent to STDOUT becomes a comment on the PR.
      echo "Cannot auto-approve test diff; more lines deleted than added."
    fi
  fi
  

All of this only scratches the surface. You can use Lint.ai to power controls for compliance (HIPAA, FISMA, SAS70, etc.), to run your test suite against every branch before it's merged, or even to fire off your deployment commands, making it a self-contained Continuous Deployment system.

So whether you're looking to run a tool like Black or Rubocop or ESLint on every commit, or frustrated by Github's low-fidelity controls on repo permissions (even with CODEOWNERS), or just in need of a simple and intuitive way to improve code quality, find out more about Lint.ai today.

---

Interested in learning more?

Limited invitations to our closed beta are available. For more information, please contact quality@lint.ai.

---

What's your linting philosophy?

I thought you'd never ask.