THIS IS F**KING INSANE i found an open source AI pentesting tool that deploys multiple agents to automate security testing. and it does more than just scan. PENTEST-SWARM-AI can: → discover subdomains and live hosts → scan ports and find vulnerabilities → validate findings with evidence → generate exportable... reports → connect to Claude Desktop and Cursor → automate scans through GitHub Actions multiple AI agents. one security testing workflow. only test systems you own or have permission to assess. REPO BELOWshow more

Here is the repo Follow @maxdeg0 for more tools like this

Don’t miss out on stuff like this Join up

this is amazing open source tool max

For real open source AI tools are taking cybersecurity to another level.

@ADanielHill

Clean setup. I'd start it from CI against a staging k3s cluster

Staging first, production later. Smart move champ

Thanks for sharing

Did you just learn about AI?

Nope Just sharing it for those that don’t know about it

Thanks, was in search of something see similar

hol, let’s see what i can do with this

Go for it champ Start with a test lab and see how the agents work together

can you tell me more about it

It uses multiple AI agents to automate recon, vulnerability scanning, and reporting. Pretty interesting how it all works together

very impressive

Damn this one is actually crazy Bro dropping alphas on weekend chad

No days off champ Glad you like it

AI agents are making security testing much easier

For real, automating the repetitive work is a game changer

Does the validate-with-evidence step actually cut false positives? That's where most scanners lose me.

Yes, it can reduce false positives by having agents verify vulnerabilities and collect evidence instead of relying on scan results alone. But it won’t eliminate them completely.

Yeah . Got that. This is Nice

pentest swarm , insane it’s replacing pen test team !

The automation is crazy, but human expertise still matters

until asi is out , for real ! agree

Facts 😅AI is moving way too fast in cybersecurity.

F**king insane indeed.

the GitHub Actions integration makes the whole thing much more practical

This is crazy. Nice alpha max

Really curious to see how well this works in real-world pentesting, especially when it comes to finding actual vulnerabilities and avoiding false positives.

想看实测，尤其是误报能压到多少。

great! thanks for sharing MAX!

Agents are great at coverage: recon, known vulns, fuzzing. Where they still struggle is business logic: refund abuse, multi-step auth flows, "is user A allowed to do this?". Best setup we've seen is agents for breadth, humans for depth.

great alpha man. the github actions part is what i'd actually use, nightly runs

First, I bookmark this. Validating findings with evidence is a part most scanners skip.

Okay I've gotta say this is pretty cool...... It's a really amazing tool

Think of it as a security team passing a finding down the line: discovery finds leads, validation checks evidence, and the report makes the result reviewable.

great share my man

Reports with reproduction steps can double as retest instructions after a fix.

Swarm 这词现在啥都能套，上次看到的还是炒股群

each one takes its own part instead of one agent doing everything in a row

evidence validation is the part most scanners skip

Wow Thank you