28 machine-enforced safety checks for AI-generated code. Catches SQL injection, hardcoded secrets, slopsquatting, reward hacking, doom loops, and 23 more failure modes. Proven by adversarial A/B testing: Keelwright Score up to 83/100.

ai-code ai-safety autonomous-coding chatgpt claude code-quality coding-agent cursor guardrails llm loop-coding owasp security vibe-coding
2 Open Issues Need Help Last updated: Jul 23, 2026

Open Issues Need Help

View All on GitHub

28 machine-enforced safety checks for AI-generated code. Catches SQL injection, hardcoded secrets, slopsquatting, reward hacking, doom loops, and 23 more failure modes. Proven by adversarial A/B testing: Keelwright Score up to 83/100.

Python
#ai-code#ai-safety#autonomous-coding#chatgpt#claude#code-quality#coding-agent#cursor#guardrails#llm#loop-coding#owasp#security#vibe-coding
good first issue

28 machine-enforced safety checks for AI-generated code. Catches SQL injection, hardcoded secrets, slopsquatting, reward hacking, doom loops, and 23 more failure modes. Proven by adversarial A/B testing: Keelwright Score up to 83/100.

Python
#ai-code#ai-safety#autonomous-coding#chatgpt#claude#code-quality#coding-agent#cursor#guardrails#llm#loop-coding#owasp#security#vibe-coding