VRead
Give agents Vercel data without a token that can wipe a project.
Feature a build or sponsor the directory. DM @pranavmore69
Pentest agents, air-gapped gadgets, compliance tooling. We skip the $68-to-$750k trading-bot posts. 40 builds on this page, each with the original post.
Give agents Vercel data without a token that can wipe a project.
Proof-of-concept code vulnerability scanner using Jev. Flags risky code without an LLM writeup.
Plugin and hooks that stop CLI agents from printing secrets. Works with Claude Code, Codex, Droid, OpenCode, and Cursor.
Local LLM gate that hides sensitive info before it reaches online model providers.
Fast Android decompiler for agents. Global xrefs in seconds, class decompile in milliseconds. Black Hat EU Arsenal.
Quantum circuit toolkit from Trail of Bits. Circuits on ECDSA.fail were built on it.
Same weights as Fable 5.1, fewer safeguards. Restricted to cyberdefenders and life scientists via trusted access.
Claude Security scans now run on Mythos 5. Public beta for Claude Enterprise. First-party.
Mythos Preview found new attacks on HAWK and a reduced AES. First-party research from Anthropic.
The full GLM-5.3 coding and cyber-defense model is now downloadable on Hugging Face. Same base as 5.2, post-trained.
Sandboxed AI harness in Zig. Landlock, eBPF, and namespaces isolate tool calls. WASM plugins over IPC.
Redact secrets and PII from logs in the browser or CLI, with stable HMAC tokens and no runtime deps.
14-agent security auditing harness in the terminal, rebuilt for Web3 on OpenCode.
GitHub Action that checks Omarchy plugins: manifests, QML lifecycle, risky capabilities, isolated install and remove.
x64 VEH Windows debugger that skips Win32 debug APIs so common anti-debug checks never see it.
Subdomain discovery for bug bounty. 2.2M+ names found in the first hours after launch.
SSH server that remote-desktops another PC straight in your terminal.
Toolkit for public Kaspa transactions, covenants, KCC20 tokens, AMMs, holders, and wallet relationships.
An advisory names the package. This tells you whether you installed the bad version while it was still live.
Reads an npm lockfile and shows if a project contains a compromised package, how it got there, and what else is at risk.
Cloudflare template for an internal static-site platform. Everything deployed is private by default.
One-click Docker setups and exploits for Nginx PoolSlip and QuicBurst, with ASLR bypass.
Move a substantial IDA 8.x infrastructure to 9.x. Open-sourced by Cellebrite Labs.
Open-source Claude agent definitions that pretty much fully automate a security audit.
Local-first trading journal. CSV import from 13 brokers, data never leaves your machine. macOS and Windows.
Encrypted clipboard that works across Mac and Windows. Search, filter, paste stack. Local memory for agents.
Local-first macOS secret manager. Link .env files, rotate keys, write them back. AES-256, Touch ID, no cloud.
AI security platform built for the new era of AI threats, defending companies against AI-powered cybercrime. YC S26.
AI agents that crawl the App Store and Play Store around the clock and deliver MASVS and GDPR digests the moment a new app ships.
Open-source AI agents that run a full pentest end to end, from recon through exploit, reverse shell and pivot, then write a client-ready PDF report.
Z.ai’s new coding and cyber-defense model, post-trained on a 743B base. Agentic coding jumps over GLM-5.2 and it is live now via GLM Coding Plan and ZCode.
Open-source CLI agent for authorized pentests. Plan, act, observe, verify, report. Local or hosted LLMs. Burp + MCP.
Open-source agent, editor and security-review tool. Headless CLI, any OpenAI-compatible endpoint.
Sends files between phones with no WiFi, Bluetooth or network. QR codes flash on one screen into the other camera at 129KB/s. Built in one night with Claude Code.
Secure enterprise-grade platform that lets companies vibe-code production software inside their own AWS environment.
Two open anti-cheat projects for Counter-Strike 2: server-side occlusion culling (CS2FOW) and a multi-module detection system (CS2AC).
This year's DEF CON badge runs open-source RISC-V silicon you can visually verify against the published RTL with an IR microscope. Doubles as a security key.
Open-source control plane for AI agents. Defense by default. Used by thousands of devs a month.
Open-source air-gapped machine (Pi 5 + fiber laser) that engraves Bitcoin seed plates in titanium.
Vibe-coded tool that transfers files from computer to phone using only the camera (completely air-gapped).
I just shipped VRead. Give agents @vercel data without giving them a token that can wipe a project. github.com/commet-labs/vr…
CLI agents love to print secrets in chat/session. So I built a secret-guard plugin prevents them via hooks and forces correct behavior, follows agent plugin standard, works with claude code, codex, droid, opencode, cursor CLIs github.com/wtfsayo/agent-…
yup, I built something similar to use local LLM to hide sensitive info from online LLM providers github.com/madeye/sealgate
Guys, I built a super fast Android decompiler called ASC. It completely replaced Jadx MCP for me and lets me analyze 10+ APKs in parallel. And this week my Agent use ASC found 2 RCE in Honor and Xiaomi! This tool now accepted by BlackHat EU Arsenal github.com/MG1937 #BHEU
Our engineer Joe Doyle built trailmix, the quantum circuit toolkit we open-sourced in June. The paper's leading circuits build on it, and his leaderboard submissions helped push the score past Google's. github.com/trailofbits/tr…
The first full paper on ECDSA.fail is on arXiv In March, Google Quantum AI reported a more efficient quantum circuit for a core step in breaking the signatures behind Bitcoin and Ethereum. It published a proof that the circuit existed and a program to verify any
Claude Fable 5.1 is available everywhere today. Claude Mythos 5.1, our model for cyberdefenders and life scientists, is available through trusted access programs. Read more: anthropic.com/claude-fable-a…
Claude Security scans now run on Claude Mythos 5, available today in public beta for all Claude Enterprise customers. Put our most capable security model to work on your codebase, no separate model access needed.
New Anthropic research: Discovering cryptographic weaknesses with Claude. Claude Mythos Preview has helped our researchers find weaknesses in cryptographic algorithms—the mathematical methods that are used to keep data private. Read more: anthropic.com/research/disco…
GLM-5.3 is now open-weight. Our most capable model for agentic coding and cyber defense is now available to download, run, and customize. Weights: huggingface.co/zai-org/GLM-5.3 Tech blog: z.ai/blog/glm-5.3
Just released a sandboxed AI harness. It's written purely in Zig and utilizes things such as landlock, eBPF, and namespaces to isolate tool calls for LLM's. It also utilizes WASM for plugins in an isolated process and fed back to the session over IPC. github.com/Midstall/chock
github.com/socprime/logto… We have just released on GitHub logtotal-sanitizer a framework-agnostic log masking tool working in-browser and CLI via Node.js. It's designed to sanitize log files before they are sent to third-party platforms for processing, analysis, or troubleshooting. Show more
🚨 Introducing AuditCode v1.0. An autonomous 14-agent security auditing harness for your terminal. Built on top of @opencode + PentestCode. Rebuilt for Web3 by @itsdarktoday & @0xscarfac3. Try it now. 🦀⚡ 🔗 github.com/itsdarktoday/a…
Omarchy plugins run as unsandboxed code inside your session. I built @OmarchyLinux Plugin CI: • manifest + entry-point checks • QML contract validation • risky capability warnings • isolated install/remove simulation Try the beta: github.com/duclucky/omarc…
About a year ago I built GhostDebug, an x64 VEH-based Windows debugger designed to bypass common debugger-detection checks by avoiding the normal Win32 debugging APIs. Check it out here: Blogpost: vollragm.github.io/posts/developi… GitHub: github.com/VollRagm/ghost…
SubMap just launched 4 hours ago. Already 2.2M+ subdomains discovered across all scans. SubMap finds more subdomains than any other tool — period. Try it free: submap.net #bugbounty #infosec
I made a SSH Server which lets you fully remote desktop into another PC with SSH, directly in your terminal! github.com/rylena/sshdesk
1/5 I’ve open-sourced kcc20-forensics, a toolkit for analyzing public Kaspa transactions, covenants, KCC20 tokens, AMMs/LPs, holders & wallet relationships. It grew out of work on KRON, Kaspa’s first covenant-based DEX. github.com/catdog1181074/…
An advisory tells you the package and the affected versions. It cannot tell you the only thing that matters on the day: whether *you* installed the bad version while it was still live. So I built something that answers that. It's called Reachable. reachable-lac.vercel.app
I built Epicenter, an open-source tool that reads an npm lockfile and shows if a project contains a compromised package, how it got there and what else may be at risk. It also replays supply-chain incidents with HydraDB Live: epicenter-frontendx.vercel.app Code: github.com/ELLA0VICTOR/ep…
wanna let your employees deploy internal apps without worrying about accidentally exposing company data? 🔒we open-sourced an internal static site platform — where everything deployed is private by default try it here: github.com/cloudflare/tem…
We’ve open-sourced one-click Docker setups and full exploits for Nginx-{PoolSlip, QuicBurst}, both with remote ASLR bypasses. Try them here: github.com/NebuSec/CyberM… PoolSlip's OOB write leaks ASLR data extremely fast, while QuicBurst's UAF takes longer.
Here goes nginx-quicburst (CVE-2026-42530), a new RCE in Nginx discovered by our security agent VEGA and demonstrated by Nebula Security. This is only the third NGINX vulnerability since 2014 to receive NGINX’s “major” severity rating. If you use Nginx 1.31 with QUIC enabled, we
If you maintain a substantial IDA infrastructure, you know the move from IDA 8.x to 9.x touched almost all of it. To ease the pain, I built ida-porter. Now open source from @CellebriteLabs: github.com/cellebrite-lab…
I open-sourced some Claude agent definitions, which pretty much fully automate a security audit: github.com/defuse/auditic…
We spent 6 months building the trading journal I actually wanted. Every other journal wanted my entire trade history — every position, every P&L, my whole edge — living in their cloud. So I built one where your data never leaves your machine. Tally is live today. 🧵
Copy on Windows. Paste on Mac. That’s it. That’s Clypy.app. 📋⚡
Tired of API keys scattered across 12 .env files So I built PassStore: a local-first secret manager for macOS. You can link to your .env files, edit on disk, pull it in, rotate in the app, write it back 🔒 AES-256-GCM, Touch ID, no cloud. Open source. v1.2.0 just shipped 🎉
Cybercrime is projected to cost the global economy $10.5 trillion every year. AI has made every company, everywhere, vulnerable The problem isn’t a lack of security tools. It’s that they were never designed to defend against AI threats We built Trident (YCS26) to change that, Show more
The most expensive minute in mobile security is the time between your app shipping and an attacker reading it. Trawlpost eliminates it — AI agents crawl the App Store and Play Store 24/7, dropping MASVS and GDPR digests on every new release the moment it lands. Live soon.
Introducing GLM-5.3: Built to Code. Ready for Cyber Defense. - Top-tier coding and agentic capabilities, achieved through post-training on the 743B base model - A major leap in cybersecurity, setting a new standard among open models Tech Blog: z.ai/blog/glm-5.3
PentesterFlow: CLI agent for AI-assisted penetration testing PentesterFlow is an open-source CLI agent that follows a plan → act → observe → verify → report cycle. It supports local and hosted LLMs through Ollama, LM Studio, Gemini, Groq, and OpenAI-compatible APIs. The Show more
Ainiux is not just an AI agent, it is a comprehensive text/code editor with AI powers, persistent chat, security review tool, supports headless CLI usage and any OpenAI API compatible endpoint. I open sourced it a week ago, been updating it ever since. github.com/petrikuittinen…
WAIT… YOU DON’T NEED WIFI TO SEND FILES ANYMORE. 🤯 No WiFi. No Bluetooth. No network. Just QR codes flashing on one phone and a camera on the other. ~129 KB/s. 2 MB in seconds. Built in one night with Claude Code. x.com/om_patel5/stat…
Today we’re partnering with AWS to launch Superblocks 3.0: the secure way for employees to vibe code production enterprise software. In a single prompt, Superblocks can replace million dollar SaaS, while IT & Security stay in control. OpenAI and Anthropic are releasing new Show more
I built two anti-cheat systems for CS2. CS2FOW is an anti-wallhack project. It uses server-side occlusion culling, so when a player is hidden behind a wall or smoke, their data is never sent to that client. A wallhack can’t render what it never receives. It also boosts FPS by Show more
The @DEFCON 34 badge packs a surprise: Andrew "bunnie" Huang's "mostly-open" Baochip-x1! hackster.io/news/the-def-c…
The ai agent control pattern is getting more validated, the reason we built prismor and used by 8k devs monthly github.com/PrismorSec/pri…
BTC Tech Nerd Army Aside from being a full-time BTC engineer, I'm a father to amazing children and a husband to an incredible wife. Family is the whole reason I'm in Bitcoin. So what happened with the Coldcard hack isn't a tech story to me. Families lost generational wealth. Show more
Vibecoded a silly little tool that transfers files from your computer to your phone air-gapped using your camera at ~50 Kbps. Nice to have when you're offline or on a plane, or need to send something super duper securely.