RAGElo: Skill for Claude Code

.agents/skills/review-pr/SKILL.md

review-pr is a skill for Claude Code, Codex from zetaalphavector/RAGElo. It costs 44 tokens per session (4,662 once invoked), scanned A, original, Apache-2.0.

A pull-request reviewer for a Python library and command-line tool that evaluates retrieval-augmented generation agents. It checks whether proposed code matches the project's behavior, design patterns, security needs, and stated changes.

In plain words
What is it for?
It is for reviewing pull requests, validating code changes, checking architecture and style consistency, and assessing whether a change is ready to merge.
Why use it?
It helps catch incomplete implementations, inconsistent code, security problems, and breaking public-interface changes before a pull request is merged.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is zetaalphavector/RAGElo's own configuration. It tells Claude Code and Codex how to work on RAGElo itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything RAGElo configures →

Reuse

Borrowing it

Nothing to install: this file belongs to zetaalphavector/RAGElo. 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/zetaalphavector/RAGElo/master/.agents/skills/review-pr/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/zetaalphavector/RAGElo

Made for: Claude Code, Codex.

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.

agentmods badge for review-pr

README.md
[![agentmods](https://agentmods.dev/badge/skills/zetaalphavector/ragelo/review-pr/github.svg)](https://agentmods.dev/skills/zetaalphavector/ragelo/review-pr)
Your own site
<a href="https://agentmods.dev/skills/zetaalphavector/ragelo/review-pr"><img src="https://agentmods.dev/badge/skills/zetaalphavector/ragelo/review-pr/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.

agentmods 80×15 button for review-pr

Your own site · 80×15
<a href="https://agentmods.dev/skills/zetaalphavector/ragelo/review-pr"><img src="https://agentmods.dev/badge/skills/zetaalphavector/ragelo/review-pr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 44 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,662 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Prompt Injection · line 26
    Subtle instructions detected that may alter agent decision-making or introduce hidden biases.
    Fix: Review content for implicit steering or bias. Ensure instructions are explicit and align with the skill's stated purpose.
  • medium Excessive Agency · line 485
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00044 $0.04662
Opus 5 $0.00022 $0.02331
Sonnet 5 $0.00009 $0.00932
Haiku 4.5 $0.00004 $0.00466

Measured 9d ago against content hash 741d742df2bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

review-pr scanned grade A with 0 findings 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/review-pr/SKILL.md · 517 lines

How it starts

The opening of the file, as written. The whole thing — 517 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Review Skill

Overview

RAGElo is a Python library and CLI for evaluating RAG (Retrieval-Augmented Generation) agents using Elo-based tournament ranking. This skill reviews pull requests to ensure they:

  • Actually implement what the PR description claims
  • Follow existing codebase patterns (the Chameleon Principle)
  • Match the code style of the existing codebase
  • Align with RAGElo's established architecture (Factory+Registry, async evaluators, Pydantic configs)
  • Address security concerns properly
  • Introduce no breaking changes to the public API (or have proper migration paths)
  • Relate to existing GitHub issues appropriately

The Chameleon Principle

The codebase should feel like it was architected by one mind, not assembled by mercenaries.

Every PR must introduce changes that blend seamlessly with existing patterns. When reviewing:

  • Find the existing pattern first - Before accepting any change, search for how similar problems are already solved
  • Reject foreign patterns - If the PR introduces a pattern that doesn't exist elsewhere, flag it ruthlessly
  • Suggest the existing way - Always recommend the established approach over novel solutions
  • Be ruthless - Pattern violations are not style nits; they are architectural debt

This principle applies to everything: architecture, code organization, naming conventions, testing approaches, and code style. AI agents are notorious for ignoring existing patterns and producing the same generic slop everywhere. The reviewer must catch this.

Output Format

The agent must produce:

  1. Summary - Brief overview of what the PR does
  2. Implementation Verification - Does the code actually do what the PR claims?
  3. Breaking Changes - Any backward-incompatible changes to the public API and migration paths
  4. Pattern & Style Conformance - Does the PR follow existing codebase patterns and style?
  5. Architecture Alignment - Is the PR consistent with RAGElo's Factory+Registry architecture?
  6. Code Quality & Testing - Test coverage, test quality, code organization
  7. Issue Linkage - Related GitHub issues
  8. Security Concerns - Issues identified against security requirements
  9. Recommendations - Required changes, suggestions, and approval status

Read the full file on GitHub · 517 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.

  1. 9d ago First seen · 517 lines · 44 tokens per session scan A 741d742df2bc

Subscribe to this mod's changes

review-pr is a skill published in the GitHub repository zetaalphavector/RAGElo (131 stars, last pushed 27d ago), licensed Apache-2.0. It adds 44 tokens to every session and 4,662 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.