pr-review-output-format

Name: pr-review-output-format
Availability: InStock
Author: Atomic

Define a standardized JSON output format for pull request review agents.

30|1|Updated Mar 15, 2026

Search Tags:#github #agents #code-review #pr-review #json-schema #review-automation #output-format

Ready to run

npx skills add https://github.com/atomicinnovation/accelerator --skill pr-review-output-format

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: pr-review-output-format
Source: https://github.com/atomicinnovation/accelerator/tree/main/skills/review/output-formats/pr-review-output-format
Command: npx skills add https://github.com/atomicinnovation/accelerator --skill pr-review-output-format

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill defines a single, machine-readable JSON output format so PR review agents produce consistent, parsable findings that an orchestrator can merge and post as inline GitHub comments.

Core Features & Use Cases

Deterministic Output: A strict JSON schema covering lens attribution, strengths, line-anchored comments, and general findings prevents ambiguous natural-language outputs.
Review-Oriented Fields: Includes severity, confidence, file path anchoring, multi-line ranges, and emoji conventions for consistent reviewer communication.
Use Case: Combine multiple agent lenses (security, correctness, performance) into harmonized PR feedback that the orchestrator can convert into GitHub review comments without manual editing.

Quick Start

Use the pr-review-output-format skill to generate a single JSON block that lists lens="security", a 2-3 sentence summary, strengths, and at least one line-anchored comment with severity and confidence.

Frequently Asked Questions about pr-review-output-format

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

FAQPage Schema

What is a standardized JSON output format for automated PR review agents?▼

A standardized JSON output format for automated PR review agents provides a strict, machine-readable schema covering lens attribution, line-anchored comments, and severity tagging. This structure prevents ambiguous natural-language outputs, ensuring orchestrators can parse and post findings as GitHub comments.

How do I structure multi-agent code review findings for GitHub inline comments?▼

To structure multi-agent code review findings for GitHub inline comments, output a strict JSON block containing lens attribution (e.g., security, correctness), a summary, strengths, and line-anchored comments. This allows an orchestrator to merge feedback from multiple agents into harmonized GitHub reviews.

Does my PR review agent need a strict JSON schema for orchestrator parsing?▼

Yes, your PR review agent needs a strict JSON schema for orchestrator parsing if you want to automate posting feedback. Enforcing JSON-only output ensures the orchestrator can reliably extract severity, confidence, and file path anchoring to generate inline GitHub comments without manual editing.

What fields are required for line-anchored comments in automated code review workflows?▼

Required fields for line-anchored comments in automated code review workflows include lens attribution, file path anchoring, multi-line ranges, severity, and confidence tagging. These review-oriented fields ensure consistent reviewer communication and precise placement of automated feedback on GitHub.

When do I need to use a JSON schema for pull request review automation?▼

You need to use a JSON schema for pull request review automation when combining multiple agent lenses, such as security and performance, into a single workflow. It specifies field-level requirements for general findings grouping and comment anchoring, enabling orchestrators to merge and post feedback seamlessly.

Can I use natural language summaries instead of JSON for multi-agent code review?▼

No, you should not use natural language summaries instead of JSON for multi-agent code review if automating the workflow. Natural language creates ambiguous outputs that orchestrators cannot parse, whereas a strict JSON schema with severity, confidence, and lens attribution enables deterministic merging and posting of GitHub comments.