Dublin startup Jentic launches a free, open source permission firewall for OpenClaw and other AI agents, giving developers credential control, fine-grained permissions, and a single killswitch.

The New Stack is a publication covering trends and technologies in cloud-native development, DevOps, and software delivery. Developers can learn about containerization, Kubernetes, and cloud computing, as well as explore topics such as microservices architecture, serverless computing, and continuous integration/continuous delivery (CI/CD) pipelines.

The New Stack

Dublin startup Jentic has launched Jentic Mini, a free, open source, self-hosted permission firewall designed to address critical security vulnerabilities in OpenClaw and other AI agents. OpenClaw, which gained 247,000 GitHub stars in 60 days, has left a security trail including 40,000+ exposed instances, credential leakage, and documented data exfiltration. Jentic Mini sits between AI agents and the APIs they call, holding credentials centrally so agents never see them, enforcing fine-grained permissions (e.g., draft but not send emails), and providing a single killswitch to cut all agent data access instantly. The product is built on 18 months of enterprise work and includes an API catalog spanning 10,000+ APIs. Jentic CEO Sean Blanchfield also predicts a broader SaaS disruption as agents increasingly build their own tools rather than relying on purchased software.

OpenClaw is a security mess. Jentic wants to fix it

A permission firewall for the agentic era