skills.rest
robotijn
robotijnCOMMUNITY
Explore all skills by this author →
GitHub

ai-code-quality-reviewer

Detect AI-specific code flaws like hallucinated imports and vacuous assertions.

3|2|Updated Jan 23, 2026
Search Tags:#code quality#code security#ai code review#llm code review#hallucinated imports#ai generated code#vacuous tests
One-click install
npx skills add https://github.com/robotijn/ctoc --skill ai-code-quality-reviewer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-code-quality-reviewer
Source: https://github.com/robotijn/ctoc/tree/main/skills/ai-quality/ai-code-quality-reviewer
Command: npx skills add https://github.com/robotijn/ctoc --skill ai-code-quality-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI-generated code from tools like Copilot, Cursor, and Claude Code often contains hidden, AI-specific flaws including hallucinated package imports, deprecated framework APIs, vacuous test assertions, and missing business rules that standard code review tools miss, leading to security vulnerabilities, technical debt, and production incidents.

Core Features & Use Cases

  • AI-Specific Pattern Detection: Flags 8 distinct AI-generation failure modes including hallucinated imports, framework version mismatches, debug prints left in production code, and unsafe concurrency patterns unique to LLM output.
  • Cross-Skill Integration: Routes findings to complementary skills like dependency-auditor for registry validation and sast-scanner for security taint analysis, avoiding duplicate work.
  • Use Case: A developer using Claude Code to generate a new .NET API endpoint can run this skill to catch any hallucinated NuGet packages, deprecated BinaryFormatter usage, or missing tenant scoping before merging the PR.

Quick Start

Invoke the ai-code-quality-reviewer skill to audit the latest AI-generated code diff for quality and security issues before merging.

Frequently Asked Questions about ai-code-quality-reviewer

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

FAQPage Schema
How do I catch hallucinated imports and vacuous tests in AI-generated code?▼

To catch hallucinated imports and vacuous tests in AI-generated code, review pull requests using an AI-specific quality reviewer that flags distinct LLM failure modes like deprecated framework APIs and unsafe concurrency patterns missed by standard linters.

What is the best way to validate AI code from Copilot or Cursor before merging?▼

The best way to validate AI code from Copilot or Cursor before merging is to run an automated pre-merge review that detects AI-specific failure modes and enforces mandatory human review gates for AI-authored production code.

Does AI code review work with TypeScript, Python, C#, and Java ecosystems?▼

AI code review works with TypeScript, Python, C#, and Java ecosystems by auditing AI-generated code diffs for framework version mismatches, missing business rules, and deprecated API patterns unique to LLM output.

Why do standard linters and SAST tools miss AI-specific code flaws?▼

Standard linters and SAST tools miss AI-specific code flaws because they are not designed to detect LLM failure modes like hallucinated package imports, vacuous test assertions, or over-engineering unique to AI-generated code.

Can I route AI code review findings to dependency and security scanners?▼

You can route AI code review findings to dependency auditors for registry validation and SAST scanners for security taint analysis, ensuring complementary tools handle cross-skill integration without duplicate work.

When should I enforce a mandatory human review gate for AI-authored code?▼

You should enforce a mandatory human review gate for AI-authored code during pull request review and pre-merge validation to satisfy automated detection requirements for AI-generation failure modes and treat warnings as bugs.

Related Skills

Explore More Software Engineering Skills→
videojsvideojs
888

review-branch

Analyzes Git/GitHub branch changes and proposes prioritized improvements for PR discussions.

Official
Intermediate
Atrium-HermesAtrium-Hermes

Vuln Scanner

Identify real security vulnerabilities in trending repositories and route disclosures.

Community
Advanced
hoanghn61hoanghn61

source-code-scanning

Scan source code and dependencies for vulnerabilities, secrets, and malicious code.

Community
Advanced
roanbrasilroanbrasil
5

static-analysis

Automate static analysis for Java, Kotlin, Python, and Rust codebases.

Community
Advanced