Sentinelle: Autonomous offensive security, powered by AI

Hey Product Hunt 👋 Solo maker here. Quick honest pitch on what this is and what it isn't. WHY I BUILT IT I was doing bug bounty on the side and burning out — not on the bugs (those are fun), but on the 4 hours of recon and scanning before I could even start thinking. So I built an AI agent to handle the boring 80%. It turned into something bigger than I expected. WHAT IT DOES You point it at a target. It does recon, vulnerability scanning, exploitation attempts, and writes a clean report. You review and submit. Fully autonomous, but you stay in control. WHAT IT'S FOUND (real findings, real targets) → OTP bypass on an auth flow → IDOR exposing editorial data across millions of records → Exposed credentials in publicly accessible sources → Cloudflare WAF evasion on protected endpoints → RBAC privilege escalation (low-priv user → admin actions) → S3 bucket misconfigurations leaking internal data Most of these are mid-to-high severity bugs that a solid hunter would find except the agent does it in 30 minutes, 24/7, in parallel on multiple targets. WHAT I'M HONEST ABOUT → It's not god-tier. It works at the level of a strong intermediate- to-senior hunter, not Orange Tsai. Yet. → Some targets it crushes, others it misses things a human would catch on instinct. → Pricing isn't finalized. PH users get extended access in exchange for honest feedback. ON THE LEGAL/ETHICAL SIDE I know "autonomous offensive AI" raises eyebrows. So I built real guardrails, not theater: → Mandatory authorization declaration per mission (user states they have permission, with framework: bug bounty program / own asset / lab / written authorization) → Hard blocklists for sensitive targets (government, healthcare, critical infrastructure, banks) → Scope validation against HackerOne / Bugcrowd / Intigriti programs → Immutable audit logs of every action → Rate limits and kill switches This isn't a tool for script kiddies. It's for people doing legitimate security work. WHAT I'M LOOKING FOR → Honest feedback from hunters and pentesters → Ideas for what to find next → Brutal criticism on what's missing or broken DMs open. Ask me anything in the comments including the hard stuff. Thanks for checking it out 🙏