review-prompt-craft

Create TOML-based negative-space review prompts for AI-assisted code reviews.

3|Updated Jan 14, 2014

Search Tags:#ci-cd #code-review #ai-review #review-workflow #context-document #review-prompt #finding-system

Ready to run

npx skills add https://github.com/prateek/dotfiles --skill review-prompt-craft

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-prompt-craft
Source: https://github.com/prateek/dotfiles/tree/main/.agents/skills/review-prompt-craft
Command: npx skills add https://github.com/prateek/dotfiles --skill review-prompt-craft

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Design and implement reliable AI-assisted code reviews by providing a structured approach to crafting negative-space review prompts, context documents, and repeatable review workflows that reduce false positives and improve findings quality.

Core Features & Use Cases

Setup-mode: Audit a project, create a TOML-based negative-space context document, and define a PR review matrix.
Review-mode: Run AI-assisted PR reviews using a single-agent or multi-agent configuration, generate prioritized, traceable findings, and drive fixes via commit messages.
Continuous improvement: Align with existing project instructions (CLAUDE.md, AGENTS.md) and tailor to deployment models and risk.
CI-ready guidance: Prepare context for PRs so automated reviews can re-run on fixes.

Quick Start

Generate a ready-to-use review context and matrix for a new PR.

Frequently Asked Questions about review-prompt-craft

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

FAQPage Schema

How do I reduce false positives in AI-assisted code reviews?▼

You reduce false positives in AI-assisted code reviews by using a structured negative-space prompt framework and a TOML-based context document to define precise review boundaries and project rules.

How do I set up an automated AI code review workflow for pull requests?▼

You set up an automated AI code review workflow by generating a TOML-based context document, defining a PR review matrix, and configuring single-agent or multi-agent pipelines for CI-ready integration.

What is a negative-space review prompt for AI code analysis?▼

A negative-space review prompt defines what AI should not flag, reducing false positives by bounding the analysis to project-specific contexts and generating prioritized, traceable findings.

Can I configure a multi-agent AI review pipeline for my software project?▼

Yes, you can configure a multi-agent AI review pipeline by defining the architecture in a TOML context file and applying the prompt framework to generate prioritized, traceable findings across agents.

Does this AI review workflow integrate with existing project instructions like CLAUDE.md?▼

Yes, the AI review workflow aligns with existing project instructions like CLAUDE.md and AGENTS.md, tailoring the generated context and review matrix to specific deployment models and risk profiles.

How do I format AI code review findings for CI-CD pipelines?▼

You format AI code review findings for CI-CD pipelines using a robust, numbered findings format that ensures traceability and allows automated reviews to re-run on subsequent fixes via commit messages.