Code Review

Name: Code Review
Availability: InStock
Author: atopile

Guide code reviews by grounding feedback in invariants and test artifacts.

3.6k|225|Updated Dec 19, 2023

Search Tags:#LLM #code-review #CI #invariants #tests #test-report.json

Ready to run

npx skills add https://github.com/atopile/atopile --skill code-review-atopile

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: Code Review
Source: https://github.com/atopile/atopile/tree/main/.claude/skills/code_review
Command: npx skills add https://github.com/atopile/atopile --skill code-review-atopile

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a structured framework for performing code reviews with a focus on invariants, tests, and verifiable feedback, enabling faster, more reliable pull-request validation.

Core Features & Use Cases

Invariant-focused review: identify and verify invariants the code must preserve and ensure changes don't violate them.
Targeted verification: propose concise tests and checks, guided by existing test artifacts (e.g., artifacts/test-report.json).
LLM-assisted analysis: helps reviewers and CI bots produce concise, actionable feedback aligned with project invariants.

Quick Start

Read the PR description and compare against the repository's PR template; run lightweight checks with:

ato dev test --llm -k <area>
ato dev compile
ato dev flags When summarizing failures, reference artifacts/test-report.json for outputs.

Frequently Asked Questions about Code Review

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

FAQPage Schema

How do I automate code review for pull requests using LLM and invariants?▼

Automated code review uses LLM analysis grounded in project invariants and test artifacts to validate pull requests. It identifies invariants the code must preserve and ensures proposed changes do not violate them, providing concise feedback for CI bots.

How do I verify code invariants and test coverage in a local developer session?▼

Verify code invariants and test coverage locally by running lightweight checks like 'ato dev test --llm' and 'ato dev compile'. Summarize validation failures by referencing verification results stored in artifacts/test-report.json.

What is invariant-focused code review and when do I need it?▼

Invariant-focused code review validates whether code changes preserve critical system rules and performance considerations. You need it for pull-request validation to ensure correctness and targeted test coverage are maintained across updates.

Can I use LLM-guided code review for CI bots and interactive local checks?▼

Yes, LLM-guided code review applies to CI bots and local developer sessions. It helps reviewers and automated systems produce actionable feedback aligned with project invariants and targeted hotspots across different environments.

What is the best way to ground LLM code review feedback in test artifacts?▼

The best way to ground LLM code review feedback is to compare changes against artifacts/test-report.json. This ensures the reviewer focuses on verifiable outputs, proposing concise tests guided by existing test artifacts for reliable validation.