What is pair-review CLAUDE.md?

Project instructions for Pair-Review, a local web app that helps people review GitHub pull requests with AI suggestions. A pull request is a proposed code change waiting for review.

Use it when working on the Pair-Review application, its Node.js and Express backend, browser interface, or local SQLite database. It helps guide features for reviewing pull requests and exchanging feedback with coding agents.

It explains how the project is structured and how its AI review partner should support human reviewers while keeping them in control. It also describes how feedback can be sent back to coding agents for another iteration.

How in-the-loop-labs/pair-review configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to in-the-loop-labs/pair-review. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/in-the-loop-labs/pair-review/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/in-the-loop-labs/pair-review

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/in-the-loop-labs/pair-review/claude-md/github.svg)](https://agentmods.dev/instructions/in-the-loop-labs/pair-review/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/in-the-loop-labs/pair-review/claude-md"><img src="https://agentmods.dev/badge/instructions/in-the-loop-labs/pair-review/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/in-the-loop-labs/pair-review/claude-md"><img src="https://agentmods.dev/badge/instructions/in-the-loop-labs/pair-review/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,784 This file is loaded in full into every session.

When invoked 3,784 The same file — it is already loaded in full.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.03784	$0.03784
Opus 5	$0.01892	$0.01892
Sonnet 5	$0.00757	$0.00757
Haiku 4.5	$0.00378	$0.00378

Measured 9d ago against content hash d738b1e550c8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

pair-review CLAUDE.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- **Tests must NEVER open browser tabs or windows.** All calls to the `open` npm package in production code are gated behind `PAIR_REVIEW_NO_OPEN` env var. Vitest config sets this globally. Any test that spawns the CLI v

CLAUDE.md · 230 lines

How it starts

The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Pair-Review Project Requirements

Overview

Pair-Review is a local web application that assists human reviewers with GitHub pull request reviews by providing AI-powered suggestions and insights. The AI acts as a pair programming partner, highlighting potential issues and noteworthy aspects to accelerate the review process while keeping the human reviewer in control.

Core Value Propositions

Tight Feedback Loop for AI Coding Agents: Enable humans to review AI-generated code and provide structured feedback back to the coding agent (Claude Code, Cursor, etc.) for iteration. This creates a continuous improvement cycle where the human stays in control while working collaboratively with AI agents.
AI-Assisted Human Review Partner: Help humans perform better code reviews by acting as a collaborative partner that highlights issues, insights, and noteworthy aspects. This is not just an automated bug finder or AI review tool - it's a partner that assists the human reviewer in making informed decisions.

Core Architecture

Backend: Node.js with Express server
Frontend: Vanilla JavaScript (no framework) - familiar GitHub-like UI
Database: SQLite for local storage of reviews and drafts
AI Integration: Claude CLI in SDK mode (claude -p for programmatic output), Antigravity CLI (agy), Codex
Distribution: npm package executable via npx pair-review

Key Design Principles

GitHub UI Familiarity: Interface should feel instantly familiar to GitHub users
Human-in-the-loop: AI suggests, human decides
Local-first: All data and processing happens locally
Progressive Enhancement: Start simple, add features incrementally
No hot-reload: Configuration is loaded once at startup. Do not flag missing cleanup on config reapply — it cannot occur at runtime.

Core Workflow

User runs npx pair-review <PR-number-or-URL>
App checks out PR branch locally (new worktree)
Web UI opens automatically in browser showing PR diff
User clicks button to trigger AI analysis (optional - simple PRs may not need it)
AI performs 3-level review analysis:
- Level 1: Analyze changes in isolation (bugs/issues in changed lines only)
- Level 2: Analyze changes in file context (consistency within file)
- Level 3: Analyze changes in codebase context (architectural consistency)
AI suggestions appear inline with code, categorized by type
User adopts/edits/discards AI suggestions and adds own comments
User submits complete review to GitHub with inline comments and overall status

Read the full file on GitHub · 230 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 230 lines · 3,784 tokens per session scan A d738b1e550c8

Subscribe to this mod's changes

pair-review CLAUDE.md is an instructions file published in the GitHub repository in-the-loop-labs/pair-review (58 stars, last pushed today), licensed Apache-2.0. It adds 3,784 tokens to every session, about $0.0189 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Code review

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

pair-review: Instructions file for Claude Code