How does the skill notify the team of its findings?

It generates a detailed Markdown report in the project directory and can optionally post a summary comment directly to the relevant GitHub issue.

Can this skill work without a formal specification file?

Yes, if a spec is missing, the skill automatically switches to analyzing code changes against the issue description and comments provided in the context.

Which AI model powers the Issue Reviewer analysis?

The skill exclusively uses the Claude Opus 4.5 model to ensure high-precision analysis of complex code changes against detailed technical specifications.

What criteria trigger a 'failure' status?

A failure is triggered if the implementation coverage is low, if critical/major bugs are identified, or if test coverage falls below the required threshold.

How does it handle very large pull requests or diffs?

The skill summarizes files over 500 lines to focus on the most impactful changes, ensuring the model remains focused on key logic rather than boilerplate.

Issue Reviewer

Name: Issue Reviewer
Author: fractary

$fractary$ byfractary

•

Collaboration Tools

Automatically audits code changes against specifications and issue descriptions to ensure implementation completeness and code quality.

The Issue Reviewer skill acts as an automated quality assurance layer for your development workflow, utilizing the advanced reasoning of Claude Opus 4.5 to verify that every requirement in a specification or issue ticket has been correctly addressed. By automatically analyzing code diffs at the start of the evaluation phase, it provides immediate, actionable feedback on implementation gaps, potential bugs, and test coverage. It categorizes results into clear status codes—Success, Warning, or Failure—to ensure that only fully compliant, high-quality code moves forward in the development lifecycle.

Key Features

01Graceful degradation to issue descriptions when formal specifications are unavailable.

022 GitHub stars

03Deep code quality analysis identifying edge cases and best practice violations.

04Automated requirement traceability between code diffs and project specifications.

05Standardized status reporting (Success/Warning/Failure) for automated workflow gating.

06Detailed Markdown report generation with specific file and line-level references.

Use Cases

01Identifying missing error handling or test cases in newly submitted code changes.

02Verifying that a complex implementation meets all acceptance criteria before human peer review.

03Automating the initial 'Evaluate' phase in autonomous AI coding agent workflows.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add fractary/claude-plugins issue-reviewer

For use in Claude.ai and ChatGPT

Download Skill