sammyboi81

AI Generated Code Risks — AI skill for Claude Code

AI community

Pitfalls specific to AI-written code - hallucinated packages, insecure defaults, over-trust - and a review routine.

How to install AI Generated Code Risks

Installs to ~/.claude/skills/sammyboi81-bulwark-mcp-ai-generated-code-risks/SKILL.md

Terminal

mkdir -p ~/.claude/skills/sammyboi81-bulwark-mcp-ai-generated-code-risks && curl -fsSL https://raw.githubusercontent.com/sammyboi81/bulwark-mcp/HEAD/skills/ai-generated-code-risks.md -o ~/.claude/skills/sammyboi81-bulwark-mcp-ai-generated-code-risks/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What AI Generated Code Risks does

name: ai-generated-code-risks description: Pitfalls specific to AI-written code - hallucinated packages, insecure defaults, over-trust - and a review routine. tags: ai, supply-chain, review

AI-written code is plausible, not verified. Research and incident reports show a meaningful share of generated code carries vulnerabilities, and models invent package names.

Slopsquatting: models hallucinate dependency names; attackers register them with malware. Before adding ANY package: confirm

Alternatives in AI

Contributing To Claude Skills Library — Thank you for your interest in contributing to the Claude Skills Library 5.3k ★
Claude Skill Web Clone — Methodology for faithfully cloning any website (static / React / WebGL) — without copying AI-hallucinated code 969 ★
Hol Guard — Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, M 485 ★