We at #BHUSA Who's around?

🚨We reported KindaRails2Shell, a critical RCE in Ruby on Rails via Active Storage. It’s not a one-shot RCE, but the preconditions are kinda common under default configurations. Patch your applications now! CVE-2026-66066 ethiack.com/info-hub/res...

KindaRails2Shell - Critical RCE in Rails via Active Storage (CVE-2026-66066) | Ethiack — Autonomous Ethical Hacking for continuous security

Ethiack research team discovered KindaRails2Shell (CVE-2026-66066): a critical RCE in Ruby on Rails via Active Storage. 500,000+ sites affected. Find out if you are and how to mitigate.

ethiack.com

Hacking With AI and Hacking AI @mohansrk.bsky.social 's HackAICon talk is live on YouTube 👀

Hacking With AI and Hacking AI: Two Sides of the Same Problem | Mohan Krishna (s1r1us) @ HackAICon

Can AI actually hack? And how easy is it to hack AI? Mohan Sriram K...

youtube.com

Are you familiar with zero-width space? This character 👉 %E2%80%8B Try inserting %E2%80%8B between characters of a blocked word. If the filter sees a different string, and the backend ignores the invisible character, you may bypass some filters.

Which AI pentesting agent is the best in the world? And how can we even measure this? (We figured it out! Keep reading 👇) The existing benchmarking methods ain't good because they work like a CTF, which doesn't resemble real-world vulns.

We just open-sourced EthiBench: a new evaluation protocol for AI pentesting agents. Ground truth and code available here: https://github.com/ethiack/ethibench

Revshells is a great tool for quickly generating shell payloads according to your needs. It supports reverse shells, bind shells, msfvenom payloads, HoaxShell, and assembled payload options.

Most recon stops at A records, and that's where some attack surface hides. Querying all DNS records allows you to expand your recon. 🧑‍💻 CNAME records can reveal third-party services and subdomain takeover opportunities 🧑‍💻 MX records expose mail infrastructure

Here's a quick way to get a bunch of JS files associated with your target.

Here's a quick one-liner for finding open redirects to chain with something more impactful. This will hit all archived URL variants, replace parameter values with your payload, and confirm live redirects using HTTPX response matching.

We've all been there: found an XSS, blocked by CSP. There's a bunch of CSP bypasses that you can try by @renniepak.nl: https://cspbypass.com It has a compilation of bypasses, based on the exact CSP you're up against.  Here’s a quick tutorial on how to use it 👇

Here's a simple concept that's helped me find a lot of bugs. Once it clicks you can't unsee it. A huge family of bugs is basically the same bug: two components read the same input and disagree about what it means.

When @jameskettle.com dropped his last-byte synchronization research back in 2019 I started hunting for race conditions. It's crazy that this still works so often 7 years later. It's such a common, widespread issue and I don't think most hackers are checking for it (usually low dupe count)

GTFOBins is a curated list of Unix binaries that may be exploited to bypass local security restrictions: SUID, sudo, capabilities, file read/write. On real targets you often land in a restricted shell, but those may help you escalate privileges, reading unauthorized files, etc

GTFOBins

GTFOBins is a curated list of Unix-like executables that can be used to bypass local security restrictions in misconfigured systems.

gtfobins.org

Hacking ≡ Magic

New redirect tricks from @castilho101 just dropped 🥷

Abusing Redirect Discrepancies to leak secrets in URLs | Ethiack — Autonomous Ethical Hacking for continuous security

Discover how attackers exploit HTTP redirect discrepancies to extract sensitive data embedded in URLs, and what you can do to prevent secret leakage in your web infrastructure.

ethiack.com

Found an IDOR vulnerability but the IDs are UUIDs? Don't drop the report yet 👇

Submitting alert(document.location) will probably get you a medium. Proving impact can make a real difference. Escalating an XSS used to be as easy as leaking cookies. That often doesn't work anymore because the HttpOnly flag is typically present on the juicy session cookies.

Are you getting a forbidden status code hitting a sensitive endpoint on a Node.js app? It may be using Fastify. The middleware may check the exact path, e.g. /internal. If the app has router normalization options enabled, the following or variations may work: //internal /internal;a=b

A lot of people are now building and using their own hackbots daily. Here's a nice blog on using AI to hunt for vulns by 0xAsm0d3us. Some takeaways that I've also been experiencing:

Claude being careful

On my way to #H121 in Lisbon 🌞 Super cool to see the first LHE from @Hacker0x01 in Portugal.

You're attacking an OAuth implementation, but it properly validates if the redirect_uri starts with https://victim.com/callback - are we cooked? Not necessarily.  We should still check if it enforces strict path matching.

It's really nice to see a lot of hackers finally coming around to the idea of using AI. Here's a cool episode from @ctbbpodcast.bsky.social on building Claude skills for hacking. Been playing with connecting Claude Code to @caido.io and loving it.

Critical Thinking - Bug Bounty Podcast

513 likes, 33 comments. "Building Claude Skills as a Bug Bounty Hunter (Ep. 166)"

youtube.com

CI/CD pipelines provide a lot of juicy attack surface. One common pattern: a build system lets you specify an output path for artifacts. If that path isn't sanitized, a ../ sequence lets you write files anywhere on the build server.

Reverse engineering MCPs like IDA Pro MCP or Ghidra MCP are really powerful to find memory corruption 0days or reversing CVEs.

Race conditions in OAuth flows can still happen in custom implementations. Tools like Turbo Intruder or even a simple multi-threaded script sending concurrent requests to the callback URL with different tokens may trigger it. Further reading here:

A couple more common OAuth 2.0 vulnerabilities

TL;DR The couple of bugs described below are common across different OAuth 2.0 implementations. The bugs may allow a malicious application to maintain an access to victim's account even after access revocation performed by the victim.

blog.avuln.com

Hackerone MCP Server (unofficial). An MCP server that gives you quick access to your HackerOne reports, programs, earnings, and scope data. Very useful tool by @OriginalSicksec for anyone that's automating but bounty hunting. GitHub repo 👇

GitHub - Sicks3c/hackerone-mcp-server: Unofficial MCP server for accessing your HackerOne reports, programs, scope, and earnings from Claude Code

Unofficial MCP server for accessing your HackerOne reports, programs, scope, and earnings from Claude Code - Sicks3c/hackerone-mcp-server

github.com

Super cool work by @s3bsrt HTTP trailers can be a blind spot. Proxies usually ignore them, but backend servers will happily merge them into the main headers, letting you sneak payloads right past security filters. Blog link 👇

Find hidden API parameters in seconds, not hours.  Arjun scans 25,890 parameter names with just 50-60 requests in under 10 seconds.

GitHub - s0md3v/Arjun: HTTP parameter discovery suite.

HTTP parameter discovery suite. Contribute to s0md3v/Arjun development by creating an account on GitHub.

github.com