AgentVerify scans your AI agent for security risks and maps findings to OWASP LLM Top 10, NIST AI RMF, and SOC 2 — then gives you a shareable verified badge so users and partners can trust what they're plugging into.

Hey Product Hunt 👋 I'm Devin, co-founder of AIBV. We built AgentVerify because AI agents are shipping everywhere right now — calling APIs, executing tools, taking actions on users' behalf — and almost none of them have been through any kind of real security review. There was no simple way to answer "is this agent safe to plug into my stack?" So we built a scanner that checks for the risks that actually matter in agentic systems, and maps every finding to frameworks people already trust — OWASP LLM Top 10, NIST AI RMF, SOC 2 — instead of inventing our own scoring system nobody recognizes. Run a scan, get a real report, and if you pass, you get a verified badge you can drop straight into your README or product page — same idea as a "tested" seal, but for AI agents. Would love feedback from this community especially — if you're building or evaluating agents, I want to know what "trustworthy" means to you and whether this actually captures it. Happy to answer anything about how the scanner works under the hood.

A real-time monitoring mode would be awesome, something that keeps checking the agent continuously instead of just running scans on demand. That way I'd catch new risks the moment a dependency update or prompt change introduces them, rather than waiting until the next manual run.

Agent Verify

The verified badge for trustworthy AI agents

The verified badge for trustworthy AI agents