
CVE-2026-126273
Differential PoC lab for CVE-2026-126273: exploits NocoDB's skipped admin bootstrap so the first unauthenticated signup gains Super Admin, and…

Differential PoC lab for CVE-2026-126273: exploits NocoDB's skipped admin bootstrap so the first unauthenticated signup gains Super Admin, and…

Library-Level eBPF Sandbox for Python (Linux & macOS): syscall-level control per module.

Node.js sandbox for running untrusted JavaScript in-process, using Proxies and controlled require to restrict built-in and external module access.

Meaningful control of data in distributed systems.

VirtualBox 7.2.20 - High - Local - File write - Guest Control copy-from DOTDOT

OPNsense 26.7.6 - High - Remote - Privilege leftover - UidField uid 0

Proof-of-concept lab for a Gitea authorization bypass where a public-only PAT with write:organization creates a private organization via POST…

ProFTPD 1.3.9d - High - Remote - Privilege leftover / File write - SFTP copy-file skips dest WRITE

Non-destructive Go verifier that checks whether a Camaleon CMS instance applies the authorization fix for CVE-2026-102261 in the media crop endpoint.

Proof-of-concept client and Docker lab reproducing CVE-2026-15583, an unauthenticated confused-deputy SSRF in Grafana MCP Server that leaks…

PoC and Docker lab for CVE-2026-49869, an unauthenticated RCE in Kestra OSS via an AuthenticationFilter path bypass that allows flow creation and…

Build guide for Red Teaming home lab. GOAD lab setup in Proxmox and pfSense, Operator/C2 and Redirectors.

Sandbox for AI coding agents. Runs Copilot CLI, Claude Code, OpenCode, Gemini CLI, Antigravity, Pi, goose or a plain shell inside a kernel-level…

Self-hosted OWASP CTF kit: one box, one free GitHub org, no cloud dependencies

Sandboxed runtime for autonomous AI agents with declarative YAML policies enforcing filesystem, network, and process constraints, plus endpoint-bound…

Gitea before 1.27.1 allows remote code execution via the diffpatch API through Git hook installation.

The exploit server for out-of-band findings. Point a target at a domain you own. Every HTTP request and every email it sends back lands in a…

Benchmark for evaluating safety risks of computer-using agents, with 104 realistic misuse scenarios across seven malicious categories, supporting…