MCP.Pizza Chef: GH05TCREW
This is a terminal app for people who test whether systems can be broken into, with permission. You describe a target and a goal, and it runs the usual scanning and probing tools for you, keeps notes, and writes up a report at the end. Extra tools can be plugged in, and it can split a big job across several helper agents. You need Python and a paid key from OpenAI, Anthropic, or a similar provider. Only use it on systems you are allowed to test.
Coding
Other
Use This MCP client To
Scan a server I own for open doors Walk me through testing my own website's login Write up a report from tonight's testing session Run a prepared testing routine against a practice target Split a large scan across several helper agents
README
GHOSTCREW FAQ
Is this legal to use?
Only against systems you own or have written permission to test; anything else breaks the law in most places.
Can I use this to check my own website for weak spots?
Yes — point it at your site, give it a goal, and it will probe and report back.
Does it need a paid key?
Yes, from OpenAI, Anthropic, or another supported provider, and web search needs a separate free one.
How hard is it to install?
You clone the project and run a setup script, or use the ready-made Docker image, so some comfort with a terminal is required.
Do I need the security tools installed first?
The Kali Docker image ships with them; otherwise it uses whatever is already on your machine.
Can I add my own tools?
Yes — it accepts extra MCP tool servers, added from inside the app with a single command.