The mypy-baseline suggest command checks if your current branch fixes enough errors from the baseline (1 by default, can be changed with the -n flag), and if not suggest you a semi-random error from the baseline to fix.

mypy-baseline suggest


The command is designed to be integrated with CI, and will do its best to correctly detect the target branch and produce stable results for the same PR.

Also, there is a native integration with GitLab CI that adds suggestions as comments to MRs. To use it:

  1. Set GITLAB_CI environment variable with an access token of the user that should be used to add the comment.

  2. Install requests: python3 -m pip install requests

  3. Run mypy-baseline suggest with --comment flag.

For example:

mypy-baseline suggest:
  # ...
    - if: "$CI_MERGE_REQUEST_ID"
    - python -m pip install mypy-baseline requests
    - git fetch origin master
    - >
      mypy-baseline suggest

With other CI providers, you can try piping the command output into reviewdog.