toxy4ny/kevlar-benchmark

Kevlar Benchmark: OWASP Top 10 for Agentic Apps (AI-Agents) 2026 a Red Team Benchmark

42
/ 100
Emerging

This tool helps cybersecurity red teams and AI security engineers find and report critical security flaws in AI agents before malicious actors do. It takes your AI agent application and automatically tests it against the OWASP Top 10 list for agent-specific injection vulnerabilities. The output is a security report detailing detected exploits and their risk levels, allowing you to strengthen your AI application's defenses.

Use this if you are responsible for the security of AI-powered applications and need to proactively identify and mitigate agent-specific injection vulnerabilities.

Not ideal if you are a software developer looking for a general-purpose testing framework not specifically focused on AI agent security.

AI Security Red Teaming Vulnerability Testing Agentic Applications Application Security
No Package No Dependents
Maintenance 10 / 25
Adoption 7 / 25
Maturity 13 / 25
Community 12 / 25

How are scores calculated?

Stars

27

Forks

4

Language

Python

License

MIT

Last pushed

Jan 20, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/toxy4ny/kevlar-benchmark"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.