Securing AI Agents in Claude With Akeyless: Demo and Setup Guide

Why AI Agents Need Runtime Authorization

AI agents across Claude, Chat, Cowork, and Claude Code are increasingly being given direct access to production databases, cloud services, and SaaS APIs, often by pasting long-lived credentials into MCP configuration files. But standing credentials are only part of the problem. Even properly authenticated agents can take unintended or destructive actions because user intent doesn’t always match LLM behavior. Authentication and access controls were designed to decide who can connect, not to govern what autonomous AI agents do after they connect.

Runtime Authorization with Akeyless

Akeyless Agentic Runtime Authority addresses both risks by ensuring the LLM never receives a credential, brokering just-in-time, least-privilege access through a customer-deployed Akeyless Gateway, and enforcing intent-aware policy at the moment of action, with full forensic traceability.

Build Your First Secure Claude Integration

Watch our solution walkthrough to see how Akeyless intercepts, evaluates, and enforces policy against a real database in a live Claude session. Once you’ve seen it in action, follow the step-by-step setup guide to get started right away. Along the way, you’ll also see how Agentic Runtime Authority complements a Claude Compliance API integration by providing runtime authorization, just-in-time access, and policy enforcement for AI agent actions.

Akeyless Agentic Runtime Authority Walkthrough and Demo

Akeyless Agentic Runtime Authority Setup guide

Akeyless Agentic Runtime Authority is available today. If you’re evaluating how to grant AI agents production access safely—without exposing credentials or losing control over autonomous actions—here’s what you need to do:

Our example will include OIDC authentication using Microsoft Entra ID and a Docker-based MySQL database as the target resource, but Akeyless supports numerous authentication methods and applications.

Setup guide overview

In this guide, we will configure the Agentic Runtime Authority as follows:

1. Create a new Akeyless account
2. Deploy the Akeyless Gateway
3. Create an OIDC Auth Method and Access Role
4. Create a dynamic secret for a MySQL database and add guardrail policies
5. Connect the Akeyless MCP to Claude
6. Use Claude to prompt the database for sensitive data

For your convenience, you can follow the step-by-step instructions in the video guide, or use the written documentation below:

In case you have any questions about the Akeyless Runtime Authority solution or configuration setup, please contact our support team.