code-review

Name: claude-plugin/code-review
Author: coderabbitai

Share on X Share on LinkedIn Share on Bluesky

100Critical

The skill executes unauthorized remote code and initiates unconstrained autonomous fix loops, creating critical risks of recursive code injection and system compromise through indirect prompt injection.

ThreatsCI SM CS BC S HITL

Threat Analysis

AI Agent Traps ↗Scanned 6/20/2026

Content InjectionPerception1 finding

Semantic ManipulationReasoningclean

Cognitive StateMemory & Learningclean

Behavioural ControlAction2 findings

SystemicMulti-Agent Dynamicsclean

Human-in-the-LoopHuman Overseerclean

Skill Info

Namecoderabbitai/claude-plugin/code-review

Registrygithub

Version2d194b823b84

PURLpkg:github/coderabbitai/claude-plugin@2d194b823b84?skill=code-review

Stars46

Installs21

Source

Repository ↗SKILL.md ↗

Install

This skill has been flagged as potentially malicious. Review the findings below before installing.

GitHub

21 installs

Skills.shDocs

npx skills add https://github.com/coderabbitai/claude-plugin

Assessments (4)

AI Agent Traps ↗

Remote code download and execution detectedcriticalcommand executionAML.T0050AML.T0053AML.T0072LLM05:2025LLM06:2025

Remote code download and execution detected

100% confidenceline 48

curl -fsSL https://cli.coderabbit.ai/install.sh | sh

Autonomous loop riskhighautonomy abuseAML.T0034.002AML.T0053LLM06:2025LLM10:2025

The skill instructs the agent to enter an 'autonomous fix-review cycle' where it modifies code and re-runs reviews without human intervention until the code is 'clean'. This creates a high risk of recursive code injection or unintended system state changes if the AI misinterprets the review output.

90% confidence

4. Fix Issues (Autonomous Workflow): 1. Implement the requested feature... 4. Fix critical and warning issues systematically 5. Re-run review to verify fixes 6. Repeat until clean

Potential for prompt injection via review outputmediumindirect prompt injection

The skill instructs the agent to process output from the CodeRabbit API. If the API or the review results are compromised, the agent will ingest these results as instructions for the 'autonomous fix' loop, leading to indirect prompt injection.

70% confidence

4. Fix Issues (Autonomous Workflow): 3. Create task list from findings 4. Fix critical and warning issues systematically

Missing allowed-tools declarationlowunauthorized tool use

Skill executes commands, writes files, or accesses the network but declares no allowed-tools, so its tool surface cannot be reviewed or constrained.

90% confidence

Undeclared network usagelowunauthorized tool use

Skill contains network access patterns (HTTP requests, socket connections) without declaring network capability in allowed-tools.

60% confidence

Missing licenseinfopolicy violation

Skill does not specify a license field. Specifying a license helps users understand usage terms.

100% confidence

Badge

Markdown

[![Mondoo Skill Check](https://mondoo.com/ai-agent-security/api/badge/github/coderabbitai/claude-plugin/code-review.svg)](https://mondoo.com/ai-agent-security/skills/github/coderabbitai/claude-plugin/code-review)

HTML

<a href="https://mondoo.com/ai-agent-security/skills/github/coderabbitai/claude-plugin/code-review"><img src="https://mondoo.com/ai-agent-security/api/badge/github/coderabbitai/claude-plugin/code-review.svg" alt="Mondoo Skill Check" /></a>

Image URL

https://mondoo.com/ai-agent-security/api/badge/github/coderabbitai/claude-plugin/code-review.svg

Secure your AI agents

Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.

Continuous scanning of skills across all registries
Policy enforcement before skills reach your agents
Integration with your existing security workflow