Which test frameworks does it support?

It is framework-agnostic and can detect configurations for Jest, Vitest, Pytest, Go testing, Mocha, and many others by scanning your repository's configuration files.

Does this skill write new tests from scratch?

No, this skill is specifically designed to review and improve existing tests. It focuses on identifying gaps and suggesting enhancements for your current test suite rather than generating new files from zero.

How does it understand the intended behavior of my code?

The skill analyzes your source code behavior and searches for implementation plans, documentation, PR descriptions, or inline comments to ensure tests align with the developer's original intent.

Can it automatically fix the issues it finds?

Yes, after performing the review, the skill will present its findings and offer to apply specific fixes and improvements directly to your test files.

Is this a replacement for a test runner?

No, this is a diagnostic and improvement tool. You should continue to use your standard test runner (like npm test or pytest) to execute the tests and verify outcomes.

Test Review & Audit Agent

Name: Test Review & Audit Agent
Author: shawn-sandy

byshawn-sandy

•

Security & Testing

Audits existing test suites to identify quality issues, coverage gaps, and misalignment with source code behavior.

The Reviewing Tests skill provides a systematic approach for auditing existing test suites against their source code. It goes beyond simple coverage metrics by analyzing implementation intent, identifying fragile mocks, detecting behavioral misalignment, and pinpointing critical untested paths. By evaluating tests across nine distinct dimensions—including naming conventions, assertion focus, and state isolation—it helps developers transition from tests that merely pass to tests that provide true confidence. The skill intelligently locates implementation plans to ensure tests align with original developer intent and offers automated fixes for identified issues.

Key Features

01Automated improvement suggestions with the ability to apply fixes directly to the codebase.

02Comprehensive audit across 9 dimensions including mock hygiene, fragility, and assertion focus.

03Deep behavioral analysis of source code to identify untested edge cases and critical paths.

04Detection of project-specific coverage targets and qualitative gap analysis.

051 GitHub stars

06Alignment verification between existing tests and developer implementation plans or design docs.

Use Cases

01Auditing legacy test suites to identify fragile tests that break during refactoring.

02Improving test maintainability by renaming vague tests and focusing assertions.

03Performing a gap analysis to ensure all branching logic and error paths are covered.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add shawn-sandy/agentics reviewing-tests

For use in Claude.ai and ChatGPT

Key Features

01Automated improvement suggestions with the ability to apply fixes directly to the codebase.

02Comprehensive audit across 9 dimensions including mock hygiene, fragility, and assertion focus.

03Deep behavioral analysis of source code to identify untested edge cases and critical paths.

04Detection of project-specific coverage targets and qualitative gap analysis.

051 GitHub stars

06Alignment verification between existing tests and developer implementation plans or design docs.

Use Cases

01Auditing legacy test suites to identify fragile tests that break during refactoring.

02Improving test maintainability by renaming vague tests and focusing assertions.

03Performing a gap analysis to ensure all branching logic and error paths are covered.