NeuroSploit: Autonomous penetration testing harness orchestrating LLMs to chain vulnerabilities

tom_doerr · x · 2026-08-15

NeuroSploit is an autonomous penetration testing framework that orchestrates a pool of LLMs to chain vulnerabilities and validates every finding with hard receipts rather than relying on model guesses. Written in Rust, it transforms targets (URLs, repos, apps, or hosts) into autonomous security engagements. It uses APIs or local models (like Claude Code, Codex, Grok) for reconnaissance, intelligently selects and runs matching agents in parallel, and chains exploits to automate offensive security operations.

Original post →

More from Infra

Infra channel →