IA × IA / Test your AI

Test your AI yourself

Your AI's vulnerabilities, at machine speed.
Version 1, deterministic: nothing here uses AI to answer, and nothing you type is stored. You describe the system; the generator builds the test plan with OWASP and MITRE ATLAS IDs, the tester agent's prompt and the report template. The AI you have already authorized does the testing, on your side. The project guides; it doesn't execute.
Defense first, attack second

If you haven't defined the system's limits yet, build the defense playbook first. Then paste the authority matrix here: the attack plan will check exactly what you promised the agent won't do.

Build your defense playbook
Fill in the form. The plan shows up here.
Take it to your AI (copy the Markdown and paste it after the prompt)ChatGPTClaudePerplexityCopilotGrok
The plan reflects what you entered and the reference frameworks. It is not an assessment of your environment or a certification. Run it only on a system you own or have explicit written authorization to test, preferably in a staging environment with test accounts.
Content provided without warranties: use it as context and validate it in your environment. Labels say what the author could verify and what is a proposal. Disclaimers and license →