AI Security,
Written in the Open.
Inkog's detection engine runs on open YAML rules that anyone can write, audit, and improve. Help us build the definitive library of AI agent security patterns.
You're early. Be among the first to shape how AI agents are secured. Your contributions today define tomorrow's standards.
Why Contribute?
Join a community shaping the future of AI security.
Transparent Defenses
Anyone can audit a rule. No black boxes. Every detection pattern is open for review, improvement, and validation by the community.
Faster Coverage
New attacks get rules in hours, not quarters. When a novel AI vulnerability emerges, the community responds immediately.
Shared Research
Your rules help secure the entire ecosystem—and your name stays attached. Build your reputation while protecting millions of agents.
Don't patch code. Patch patterns.
Writing a rule is the fastest way to turn your red teaming knowledge into protection for the entire AI ecosystem.
Each YAML rule you contribute gets deployed to thousands of CI/CD pipelines, protecting agents before they ever reach production.
1id: detect-prompt-injection 2name: Prompt Injection via User Input 3severity: HIGH 4frameworks: 5 - langchain 6 - crewai 7pattern: 8 - prompt.contains(user_input) 9 - not: input.sanitized10message: "User input flows directly into prompt without sanitization"11remediation: "Use input validation and prompt templating"12cwe: CWE-9413owasp: LLM01@agent_researcher
Joined Oct 2024
Join the Elite.
Contributors get permanent attribution, exclusive swag, and "Agent Researcher" status in our community.
- Tiered researcher levels (1-5) based on contributions
- Leaderboard recognition and Hall of Fame
- Exclusive swag and conference invitations
- Your name in every rule you contribute
Ready to make AI agents safer?
Join us in building the foundation for AI agent security. Every contribution matters.