skills.rest
getlien
getlienOFFICIAL
Explore all skills by this author →
GitHub

dogfood-veille

Create pull requests with intentional code violations to evaluate AI review quality.

2|Updated Nov 20, 2025
Search Tags:#code quality#testing#ci/cd#code review#ai evaluation#dogfooding
One-click install
npx skills add https://github.com/getlien/lien --skill dogfood-veille
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dogfood-veille
Source: https://github.com/getlien/lien/tree/main/.claude/skills/dogfood-veille
Command: npx skills add https://github.com/getlien/lien --skill dogfood-veille

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the evaluation of AI-driven code review systems by creating controlled test cases with known code quality violations.

Core Features & Use Cases

  • Automated Test Case Generation: Creates pull requests with specific code smells (complexity, DRY, SOLID, KISS, coupling).
  • Review Quality Scoring: Fetches AI review comments and compares them against ground truth to measure detection accuracy, metric accuracy, and comment quality.
  • Use Case: Developers can use this Skill to continuously monitor and improve the effectiveness of their AI code review tools, ensuring they accurately identify and report code quality issues.

Quick Start

Run the dogfood-veille skill to evaluate the AI review system by creating a PR with known violations.

Frequently Asked Questions about dogfood-veille

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

FAQPage Schema
How do I evaluate AI code review quality using pull requests?▼

Evaluating AI code review quality involves programmatically creating pull requests with intentional code violations, fetching the AI's review comments, and comparing them against ground truth to calculate detection rate, metric accuracy, and architectural adherence.

What code quality issues can I test against an automated code review system?▼

You can test AI code review systems against specific code smells including complexity, DRY, SOLID, KISS, and coupling issues to measure how accurately the AI identifies and reports these architectural violations.

Does this automated code review evaluation process require GitHub CLI?▼

Yes, evaluating AI code review quality programmatically requires GitHub CLI interactions and Git operations to create test pull requests and fetch the AI review comments for detailed parsing and scoring.

What is the best way to measure the detection rate of an AI code reviewer?▼

The best way to measure detection rate is by injecting known code violations into test pull requests, parsing the AI review comments, and comparing them against predefined ground truth metrics to calculate detection accuracy.

How do I continuously monitor code review tools for SOLID and DRY violations?▼

You can continuously monitor AI code review tools by running an evaluation process that generates pull requests with specific code smells like SOLID and DRY violations, then scores the AI's detection and reporting accuracy against ground truth.

Related Skills

Explore More Software Engineering Skills→
sourcegraphsourcegraph
32

mine-tasks

Mine merged pull requests to generate benchmark tasks with Docker environments.

Official
Advanced
letta-ailetta-ai
134

nginx-configuration

Configure Nginx logging, rate limiting, and custom error pages.

Official
Advanced
pacaplanpacaplan
30

flokay:finalize-pr

Push a pull request, monitor CI status, and iteratively fix failures.

Community
Advanced
iani-kuliiani-kuli

github-code-review

Analyzes git diffs and submits GitHub pull request reviews via CLI.

Community
Advanced