AI Engineer · October 28, 2026

Agent authorization: broker credentials outside the runtime and test gateway bypass

video Agent Security AI Engineering

Agent authorization: broker credentials outside the runtime and test gateway bypass video thumbnail

Why it matters

Kamil Potrec separates the user, agent and downstream service identities in a delegated request. His design uses scoped credentials and an external broker or gateway to inject secrets and enforce policy. It also identifies the limits: direct network paths can bypass the gateway, local tools remain outside it, and the broker becomes a sensitive dependency.

My takeaway: Map each hop from user to agent to service, recording who delegated what. Keep downstream credentials outside the agent runtime and validate caller identity, issuer, scope and policy at the service boundary. Test revocation, denied operations and direct-egress bypasses. Audit both user and agent identity, and assess local tool permissions separately.

Keep exploring

More curated notes connected through Agent Security and AI Engineering.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.

OECD.AI Wonk · guide

A five-step roadmap to closing the AI evaluation gap

The roadmap addresses evaluation results that overstate real-world performance or fail to transfer across deployment contexts. Its five steps balance standardized and local tests, evaluate throughout the lifecycle, build qualified assurance and communication capacity, tailor tests to each value-chain actor and technology, and use a coordinated, trusted process for updating methods.

OWASP GenAI Security Project · tool

FinBot CTF Is Live: A Hands-On Companion to the OWASP GenAI Security Project

OWASP FinBot is a hands-on agentic-security CTF built around a simulated multi-agent financial-services platform with real tool access. Its challenges cover prompt injection, tool misuse, policy bypass, data exfiltration, privilege escalation, remote code execution, shared context, and compromised MCP servers.