quorum-cli

Automate code reviews with local AST analysis and LLM reasoning.

2|Updated Mar 24, 2026
One-click install
npx skills add https://github.com/jsnyder/quorum --skill quorum-cli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quorum-cli
Source: https://github.com/jsnyder/quorum/tree/main/skills/quorum-cli
Command: npx skills add https://github.com/jsnyder/quorum --skill quorum-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires quorum, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

quorum-cli simplifies and enhances the code review process by integrating local AST analysis, LLM review capabilities, and a variety of modes tailored for different use cases.

Core Features & Use Cases

  • Multi-Source Code Review: Combines local AST analysis with LLM cold read, linter orchestration, and ast-grep rules for comprehensive reviews.
  • Feedback-Calibrated Findings: Uses your feedback history to refine and calibrate findings for better accuracy.
  • Flexible Review Modes: Offers local-only, LLM-augmented, parallel, compact, ensemble, and daemon modes to fit different review scenarios.

Quick Start

Run the following command to start a review:

quorum review src/auth.py

Frequently Asked Questions about quorum-cli

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

FAQPage Schema
How do I automate code review with AST analysis?

Automate code review by running local AST analysis combined with ast-grep rules and linter orchestration to parse code structure and identify issues without relying solely on manual inspection.

How do I use LLM reasoning for software quality checks?

Use LLM reasoning for software quality checks by configuring an OpenAI-compatible endpoint and API key to perform cold read reviews of source files alongside local linter orchestration.

Do I need an API key to run LLM-augmented code review?

Yes, you need a configured API key and an OpenAI-compatible endpoint to run LLM-augmented code review. You can alternatively use local-only modes if you prefer to avoid external API dependencies.

What's the best way to improve linter accuracy with feedback history?

The best way to improve linter accuracy is to use feedback-calibrated findings, where your historical review feedback refines AST analysis and LLM reasoning to reduce false positives over time.

What are the limitations of local-only code review modes?

Local-only code review modes limit analysis to AST parsing and linter orchestration, missing the deeper semantic context and complex logic flaws that LLM-based reasoning can identify.