review

Automate structured code reviews and generate consolidated findings reports.

16|2|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/Kibertum/tausik-core --skill review-kibertum
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review
Source: https://github.com/Kibertum/tausik-core/tree/main/agents/skills/review
Command: npx skills add https://github.com/Kibertum/tausik-core --skill review-kibertum

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill guides rigorous code review processes, helping teams detect bugs, vulnerabilities, and architectural antipatterns before changes ship.

Core Features & Use Cases

  • Structured Review Workflow: loads task context, stack conventions, and project rules to ensure comprehensive checks.
  • Multi-Phase Review: uses separate agents to assess quality, implementation, tests, simplification, and documentation, plus an adversarial critic.
  • Deterministic Outputs: aggregates findings into a consolidated report, supporting traceability and gating.
  • Use Case: when a developer asks to review a PR, the Skill walks through the code, flags issues, and suggests fixes aligned with acceptance criteria.

Quick Start

Submit a PR for review and run /review on the target file or directory to generate a structured analysis.

Frequently Asked Questions about review

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

FAQPage Schema
What is an automated structured code review and how does it identify bugs?

An automated structured code review loads task context and stack conventions to identify defects and improvements. It applies multi-phase checks across quality, implementation, testing, simplification, and documentation to flag issues and suggest fixes.

How do I run a code review on a pull request to catch security vulnerabilities?

To review a pull request, submit the PR and run the review command on the target file or directory. The process analyzes code changes against acceptance criteria to catch security vulnerabilities and generates a consolidated findings report.

Can I use this code review workflow for changes across multiple files and directories?

Yes, you can apply this code review workflow to changes across multiple files or directories. It loads project rules and stack conventions to perform comprehensive checks across the entire codebase or targeted directories.

What's the best way to ensure my code changes meet acceptance criteria before shipping?

The best way to verify code changes meet acceptance criteria is running a multi-phase review with an adversarial critic. It assesses quality, implementation, tests, and documentation, aggregating findings into a consolidated report for traceability.

Do I need a task with acceptance criteria to perform a code review?

Yes, a task with acceptance criteria and a PR or codebase to review is required. The review process uses these acceptance criteria to align suggested fixes and ensure the code changes meet the defined project requirements.

Why does my code review output include findings from parallel agents?

Your code review output includes findings from parallel agents because the process uses separate agents to assess quality, implementation, tests, simplification, and documentation. This multi-phase approach ensures rigorous defect detection and architectural analysis.