Developersummit
  • HOME
  • SPEAKERS
  • SESSIONS
  • SCHEDULE
  • BUY TICKETS
  • FAQ
  • CONTACT
  • ONDEMAND
saltmarch

GIDS news media, articles, insights and virtual events educate and illuminate its audiences so they can be fully prepared to deal with the new realities at work and in their professions.

Saltmarch On-Demand
Media

Our Experts

Videos On Demand

Insights

Connect

About Us

Privacy Policy

Code of Conduct

Terms & Conditions

Contact Us

Sponsor GIDS 2027

Subscribe to Developersummit

Get the latest event updates, and insights from today's leading voices.

© 2026-2027 Saltmarch. All rights reserved.

Who Is This Agent Acting For? Authorization for MCP and Agent Tools
RegisterTwitterLinkedInFacebook

< session />

Who Is This Agent Acting For? Authorization for MCP and Agent Tools

Thu, April 29Reliability, Observability & SecurityAI-Native SoftwareBackend Engineering & Runtimes

When an AI agent calls a tool, someone’s authority is being used. But whose? The user who asked, the agent itself, or the platform running it? Get that wrong, and a helpful assistant can become a confused deputy with access to far more than intended.

In this session, you will explore how the Model Context Protocol addresses this problem through its authorization specification, including changes introduced in the July 2026 revision. You will walk through the complete authorization flow step by step, including OAuth 2.1 with PKCE, protected resource metadata, resource indicators that prevent tokens from being replayed against the wrong server, and issuer validation. You will also examine the move to stateless MCP, the deprecation of Dynamic Client Registration in favour of Client ID Metadata Documents, and tighter requirements around issuers and credential binding.

The session then moves beyond the authorization handshake to a decision the specification cannot make for you: whether a particular tool call, with its particular arguments, should actually be allowed. You will leave with a clear mental model, a reusable sequence diagram of the complete flow, and a checklist for building MCP servers that know exactly who they are working for.

What You Will Learn

  • How authorization works across users, agents, platforms, and MCP servers

  • How OAuth 2.1 with PKCE, protected resource metadata, resource indicators, and issuer validation fit together

  • What the July 2026 MCP authorization changes mean for identity and trust

  • How to reason about whether a particular agent tool call and its arguments should be authorised

Who Should Attend

AI Engineers, Security Engineers, Platform Engineers, Software Architects, Backend Developers, and practitioners building MCP servers or agent systems that invoke external tools.

< speaker_info />

About the speaker

Brent Laster

Brent Laster

Author & AI Software Engineering Educator

Brent Laster is a global trainer, author, and speaker on open-source technologies, as well as an experienced developer, manager, and director. He is also the founder and president of Tech Skills Transformations, LLC – a company dedicated to making technology understandable and usable. Throughout his career in software development and management, Brent has always made time to learn and develop both technical and leadership skills and share them with others. He believes that regardless of the topic or technology, there’s no substitute for the excitement and sense of potential that come from providing others with the knowledge they need to accomplish their goals.

Related Talks

Contain the Blast Radius of AI-Accelerated Attacks

Fri, April 30

Contain the Blast Radius of AI-Accelerated Attacks

Rohit Bhardwaj
Your Prompt Is Not a Sandbox: Engineering Lessons from the 2026 Agent Breaches

Tue, April 27

Your Prompt Is Not a Sandbox: Engineering Lessons from the 2026 Agent Breaches

Brent Laster
Testing in Production: No Kidding

Thu, April 29

Testing in Production: No Kidding

Venkat Subramaniam

On-Demand Talks

All On-Demand »