What is secorizonai?

SecorizonAI is a command line AI assistant built specifically for penetration testers, security professionals who are hired to find vulnerabilities in systems they are authorized to test. Instead of an AI that suggests commands for you to type, this one actually executes shell commands: you describe a task, the AI plans and runs the steps, reads the output, and continues until the job is done. The core loop is simple: your message goes to a local AI model, the model replies with text for you plus an optional shell command to run or a web search to perform, the tool executes those actions and feeds the results back, and the cycle repeats until the model says it is finished. A key design decision is that all AI inference, the process of running the model, happens locally on your own hardware via Ollama, a tool for running AI language models offline. No data is sent to a cloud provider, and no API key is required for the main AI functionality. Built in web search is included for fetching current vulnerability advisories or recent write ups, and it also requires no API key. The entire tool compiles to a single Go binary with one dependency, making it easy to install. The system prompt, the set of instructions that shape how the AI behaves, is a plain Markdown file you can edit and replace. The repository ships a skeleton example. Detailed pentesting playbooks covering reconnaissance, web testing, exploit development, and Active Directory assessment are part of a paid Pro offering rather than the open source package. You would use this if you want an AI assistant that runs on your own machine and can execute authorized security testing tasks directly rather than only advising. Running it inside Docker is recommended so the agent's commands stay sandboxed from your main filesystem, and the project notes it has no built in authentication of its own.