Replying to @robiartec @spacexai
Shannon runs real exploits (creates users, submits forms, changes data), so it should only be used in a staging environment.
You give it test credentials. Set the login URL, username, and password in the config, and describe the login flow in plain steps.
We built Shannon as a tool for human pentesters. It handles exploit-verified testing so experts can spend their time on the more complex chained-attack work.
That part is on Grok’s side: its guardrails work with pentesting out of the box. If you’d rather use Anthropic or OpenAI, complete their verification and Shannon runs on those too.
Replying to @sergiorocaa @spacexai
Usually 1 to 1.5 hours, depending on app size and model!
Yes. There’s an official GitHub Action and a GitLab CI component. It can gate releases on proven Critical and High findings. One note: CI uses an xAI API key rather than the subscription.
Replying to @angeldot_ @spacexai
Thank you! Its fully open source, and you provide the llm model of your choice!
Let us know what it finds! If you run into any issues feel free to ask for support in our community Discord: discord.gg/a4CFDyKBB
Replying to @nattcornelis @spacexai
That’s the goal! Pentesting should not be a $4,000 engagement.
Grok is now your on-demand AI hacker.
Connect your xAI subscription to Shannon and run full pentests on your own apps.
No separate API bill.
No cyber verification required.
Every finding ships with a working proof of concept.
Open source, 49K+ stars.
@spacexai
6/ So if you already subscribe to Grok:
You may already have most of what you need for an on-demand AI pentester.
Shannon is open source:
github.com/KeygraphHQ/shanno…
Grok + Shannon is a pretty powerful combination.
