APort

First Passport + Guardrail for AI Agents

4 followers

First Passport + Guardrail for AI Agents

4 followers

AI agent guardrails and authorization infrastructure for production AI. APort provides runtime guardrails, AI passport verification, and policy enforcement before the API call. Secure OpenClaw, LangChain, CrewAI agents with pre-action verification and court-admissible proof.

Free

Launch tags:Artificial Intelligence•GitHub•YC Application

Launch Team

CloudflareStill paying to deploy your apps? Try Cloudflare Workers for free

Promoted

Unispend

Maker

📌

We completed a CTF and after over 3000 attempts, the AI agents were compromised by APort guardrails protected funds from being withdrawn. We know that AI can be prompt injected, hacked or convinced to take unintended actions. APort helps you stop Your AI Agent Before It Does Something Stupid APort is different from other Guardrails. It is not content filtering after - authorization before. The numbers and data speaks for itself. See our CTF Leaderboard (https://vault.aport.io/leaderboa...) where Level 5 with APort guardrails has not been cracked even when the AI agents was compromised over 3,000 times

Report

8mo ago

Reviews