test-review

Review test, e2e, and CI changes to detect theater tests and workflow misconfiguration.

3|Updated Jul 2, 2026
One-click install
npx skills add https://github.com/ArangoGutierrez/claude-toolkit --skill test-review-arangogutierrez
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: test-review
Source: https://github.com/ArangoGutierrez/claude-toolkit/tree/main/.claude/skills/test-review
Command: npx skills add https://github.com/ArangoGutierrez/claude-toolkit --skill test-review-arangogutierrez

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Tests that pass even when the code they cover is broken give false confidence, and CI misconfigurations silently weaken security and gating. This Skill reviews test, e2e, and CI changes to expose theater tests, flaky e2e patterns, and GitHub Actions or Prow misconfiguration before they reach human review. ## Core Features & Use Cases - Theater-test detection: Applies the deletion test (would the test fail if the code under test were deleted?) plus checks for tautological assertions, over-mocking, and guard fixtures that cannot trip the guard. - e2e and integration quality review: Flags sleep-based synchronization, retry-masked failures, missing teardown, order dependence, and non-hermetic network dependencies. - CI config review: Checks GitHub Actions for unpinned actions, over-broad permissions, pull_request_target risks, and shell injection, plus Prow jobs for regex trigger mismatches and missing resource limits. - Use Case: Before requesting review on a PR that adds tests and a new workflow, run the review to get each finding as file:line with a category (theater-test/flakiness/ci-config/coverage-gap) and severity (must-fix/should-fix/consider). ## Quick Start Ask the assistant to review these tests and the workflow changes in the current diff for theater tests and CI misconfiguration.

Frequently Asked Questions about test-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect theater tests in a pull request?▼

Apply the deletion test: mentally delete the code under test and check whether the test still passes. A test that survives deletion asserts nothing. Also check for tautological assertions, expected values re-derived from the implementation, and over-mocking.

How to review GitHub Actions workflows for security issues?▼

Check that actions are pinned to full commit SHAs, permissions are least-privilege, pull_request_target never executes fork code, and untrusted github.event values are passed through env variables rather than inlined into run steps. Running actionlint sharpens the review.

Does this test review require actionlint or Prow checkconfig?▼

No, both are optional. The review runs actionlint or checkconfig when present on PATH and notes their absence when a repo ships CI without them, but the checklist-based review works without either tool.

What makes an e2e test flaky and how is it flagged?▼

Common flakiness signals include fixed sleep calls instead of polling conditions, blanket retries masking real failures, order or state dependence between tests, index-based access into unordered lists, and timeouts tuned for fast dev machines rather than CI runners.

When should I not use test review on a change?▼

Do not use it for general code correctness review, which belongs to a language-specific review pass, or for style nits that a formatter or linter already handles. It also does not demand full coverage, only coverage of behavior the PR claims to add.