Prompt Fuzzer:
GenAI Apps Vulnerability Assessment

Test and harden the system prompt of your GenAI apps with Prompt Fuzzer!  

Start fuzzing with this interactive, open source tool that simulates LLM-based attacks such as sophisticated prompt injections, system prompt leaks, and harmful content elicitations, and gives a security evaluation and score.

Background Sphere

How does the Prompt Fuzzer work?

Get the Prompt Fuzzer from GitHub


Simply go to the repository and start running the Prompt Fuzzer. Want to add more content to the Fuzzer? All contributions are welcome!

Start testing your system prompt


The tool will run various dynamic llm-based attacks and return a security score and evaluation based on the outcomes of the tests

Test yourself with the Playground!


Iterate the settings on a chat format and harden your system prompt as many times as you like until it's safe and secure

Take a quick glance


Check out the project on GitHub

As easy as 1, 2, 3. Get the Prompt Fuzzer today and start securing your GenAI apps