mintwork

mintwork / honest-tests

Catch AI-written tests that can't fail

Your AI wrote 200 tests. Many of them can't fail. honest-tests finds them, and proves it by breaking your code and watching the tests stay green.

npx @mintwork/honest-tests audit .               # find tests that can't fail
npx @mintwork/honest-tests mutate src/pricing.ts # prove it: break the code, see who notices
honest-tests audit finds 11 of 15 tests that can't fail in a sample project; honest-tests mutate then shows 22 of 26 code changes survive the test suite

The problem

Ask an agent to "add tests" and you get a wall of green checkmarks. Look closer: a value compared to itself, a test with no assertion, an assertion inside a try/catch that swallows the failure, a test that checks the mock it just configured. Every one passes. Coverage goes up. Nothing is verified.

honest-tests is a skill that stops Claude Code, Codex and Cursor from writing these tests, plus a CLI that finds them in the suite you already have.

What it does

Install

Claude Code (plugin with skill and slash commands):

/plugin marketplace add mintwork/honest-tests
/plugin install honest-tests@honest-tests

Codex and Cursor:

npx @mintwork/honest-tests init codex    # adds a section to AGENTS.md
npx @mintwork/honest-tests init cursor   # writes .cursor/rules/honest-tests.mdc

GitHub Action

Comment on every pull request with what its tests are worth:

- uses: actions/checkout@v7
  with:
    fetch-depth: 0
- uses: mintwork/honest-tests@v0

Open source, MIT licensed

Built in CI and published to npm with signed provenance. The full docs, changelog and issues are on GitHub.