
agentseal
Security toolkit for AI agents. Scan your machine for dangerous skills and MCP configs, monitor for supply chain attacks, test prompt injection…

Security toolkit for AI agents. Scan your machine for dangerous skills and MCP configs, monitor for supply chain attacks, test prompt injection…

Access control for AI agents. Set what Claude Code, Codex, Gemini, Cursor and any MCP server are allowed to do, review risky actions before they run,…

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

Signing-key abuse and update exploitation framework

Security benchmark for evaluating OpenClaw agents against adversarial execution contexts including poisoned files, injected skills, misleading tool…

Static security scanner for AI agent skill packages. Detects malicious SKILL.md files and bundled scripts before they run.

End-to-end simulation of a Python dependency confusion attack, sudo privilege escalation (CVE-2025-32463), and rootkit-based persistence - with full…

Detects CVE-2026-45321 (TanStack supply chain compromise) and Mini Shai-Hulud worm artifacts. Scans node_modules, lockfiles, persistence hooks…

PoC for CVE-2025-54416 tj-actions/branch-names command injection

CVE-2024-24590 ClearML RCE&CMD POC

CVE-2025-53547 one of poc code

Azure IoT Hub where exposure of an owner-level Shared Access Key enables unauthenticated remote code execution (RCE) against connected IoT devices.…

CocoaPods RCE Vulnerability CVE-2024-38366

LlamaStack-RCE: Deterministic Supply Chain Exploitation & Hardening Framework [CVE-2024-50050] Focus on AI Security Research…

A CI/CD Red Team Framework for demonstrating Build Pipeline security risks.

LD_PRELOAD-based tool that hijacks gcc to inject malicious code into binaries during linking, enabling stealthy backdoor deployment without source…

CVE-2026-67595 — Embedded malicious JavaScript (spyware) in VaahCMS 2.0.0–2.3.4 official releases. CVSS 8.1. Advisory + detection.

Layer-2 supply-chain hardening for MCP servers — Ed25519-signed tool manifests, runtime spawn-attestation, default-deny argument sanitizer. Defends…