test-challenger

Challenge unit tests to detect false positives from implementation-derived oracles.

7|2|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/AndurilCode/craftwork --skill test-challenger
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-challenger
Source: https://github.com/AndurilCode/craftwork/tree/main/skills/test-challenger
Command: npx skills add https://github.com/AndurilCode/craftwork --skill test-challenger

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It challenges unit tests that appear to “pass” while secretly validating the implementation instead of the intended specification, helping you avoid false confidence from brittle or tautological tests.

Core Features & Use Cases

  • False-positive test detection: Identifies patterns where the oracle comes from running the code rather than independent domain truth.
  • Verdict-driven auditing: Classifies tests as Solid, Suspicious, Likely False Positive, or Vacuous based on a catalog of challenge patterns.
  • Test remediation guidance: Recommends fixes such as independent oracle rewrites, boundary/negative test additions, property-based testing, or deletion of vacuous tests.
  • Use case: Review AI-generated or coverage-driven tests that increased line coverage but may be mirroring bugs instead of verifying correct behavior.

Quick Start

Ask the skill to challenge your tests by providing the test file(s) and the source file(s) they are intended to validate, and request a pattern-based verdict plus remediation steps.

Frequently Asked Questions about test-challenger

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

FAQPage Schema
How do I detect false positives in AI-generated unit tests?

Detect false-positive unit tests by challenging the test suite against the source code to identify oracles derived from the implementation. This auditing process classifies tests as Solid, Suspicious, Likely False Positive, or Vacuous using a catalog of challenge patterns.

Why does my test coverage increase but tests pass despite wrong behavior?

Tests pass despite wrong behavior because coverage-driven tests often derive their oracle from running the implementation rather than the specification. Auditing these tests reveals tautological patterns that validate the code instead of verifying correct behavior.

How do I audit code-coverage-driven tests for implementation mirroring?

Audit coverage-driven tests by providing the test files and corresponding source files to challenge against a false-positive pattern catalog. This produces a per-test verdict and recommends concrete remediation like independent-oracle rewrites and boundary test coverage.

What is the best way to fix vacuous or tautological unit tests?

Fix vacuous or tautological unit tests by applying remediation guidance such as rewriting with independent oracles, adding boundary or negative test coverage, introducing property-based testing, or deleting vacuous tests entirely.

Can I review a test suite to check if it validates spec vs implementation?

Review a test suite to check spec vs implementation validation by challenging the tests to detect false-positive patterns. This highlights highest-risk findings where the oracle comes from running the code rather than independent domain truth.