agentsclimarketplace
Sign in

Generate tests

Skill axellab/adversarial-testing-toolkit/skills/generate-tests

Generate tests for a target (file, feature, or change), matching the project's own framework and conventions, then run them and report results with evidence. Use when the user says "write tests for X", "generá y corré los tests", "add test coverage", or wants actual executable tests — not just recommendations. Detects the stack automatically and never fakes a passing run.From its SKILL.md

Install
npx -y skills add axellab/adversarial-testing-toolkit --skill generate-tests

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.9 KB, as published. Nobody here has run it

Generate and run tests

This skill produces real, executable tests and runs them, delegating to the test-engineer agent. It composes naturally after /test-plan.

What to do

  1. Identify the target (file, feature, module, or current change). If a prior /test-plan / weakpoint-scout report exists in the conversation, pass that plan through so generation targets the ranked weak points. Otherwise the agent will identify cases itself.
  2. Delegate to test-engineer. It will:
    • Detect the language, test framework, real test command, and existing conventions.
    • Generate tests that fit the repo (same runner, style, placement, mocks).
    • Run them and report the real command + output.
  3. Return its results. Present what was generated, the execution evidence, any bugs the tests surfaced (as findings), and remaining coverage gaps.

Guardrails

  • Never fake a run. If the tests couldn't be executed (missing deps, unknown command), the report must say they were generated but NOT run, with the exact command to run or a request to install what's needed. Don't claim green when you don't know.
  • Fit in, don't take over. No new test framework or dependency without asking. Keep tests isolated from real networks/DBs/paid APIs unless the repo already does otherwise.
  • A failing generated test may be a real bug, not a mistake — surface it as a finding rather than quietly "fixing" it to pass.
  • Respond in the user's language.

Keep looking

Skills are one crate of 626,474. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.