software-test-review

Review TDD Red-phase tests against acceptance criteria and codebase conventions.

902|58|Updated May 31, 2012
One-click install
npx skills add https://github.com/stencila/stencila --skill software-test-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: software-test-review
Source: https://github.com/stencila/stencila/tree/main/.stencila/skills/software-test-review
Command: npx skills add https://github.com/stencila/stencila --skill software-test-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Tests written during the Red phase of test-driven development often have coverage gaps, convention violations, or trivial assertions that inflate test counts without catching real bugs. This Skill evaluates those tests before they drive implementation, producing a structured Accept or Revise recommendation.

Core Features & Use Cases

  • Acceptance-Criteria Coverage Matrix: Maps each acceptance criterion to the tests that verify it, flagging missing, weak, or non-test-deliverable criteria.
  • Convention and Quality Review: Discovers existing codebase test conventions independently and evaluates naming, assertions, isolation, readability, and minimality.
  • Trivial Test Detection: Identifies tautological assertions, constructor echo tests, and type-system-verified checks that add maintenance cost without testing value.
  • Use Case: After a test-creation agent writes Red-phase tests for a new slice, run this review to confirm every acceptance criterion is covered and failures indicate correctly missing implementation before starting the Green phase.

Quick Start

Review the Red-phase tests in tests/test_parser.py against the acceptance criteria for the CSV parser slice and tell me whether to accept or revise them.

Frequently Asked Questions about software-test-review

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

FAQPage Schema
How do I review TDD Red-phase tests before implementation?

Provide the slice name, acceptance criteria, target packages, and test file paths. The review checks coverage of each criterion, convention conformance, test quality, and whether failures indicate correctly missing implementation, then recommends Accept or Revise.

What makes a test trivial or low-value in a test suite?

Trivial tests include tautological assertions, constructor echo tests, getter/setter round-trips with no logic, and checks the compiler or type system already guarantees. They add maintenance cost without catching real bugs and are flagged for removal.

Does this skill write or modify test files?

No. It only reads test files and existing codebase tests, evaluates them, and produces a structured review report. All fixes are left to the test-creation agent based on the actionable findings.

Can the review run without test execution results?

Yes. Without execution results, the review assesses coverage and quality dimensions alone and skips Red-phase failure validation, noting the limitation in the report. Accept or Revise recommendations are still produced.

When should tests be accepted versus revised in TDD?

Accept when all test-appropriate acceptance criteria are covered, no High-severity findings exist, and Red-phase failures indicate missing implementation. Revise when coverage is missing, test bugs or syntax errors exist, or the suite is majority trivial tests.