The Model Context Protocol (MCP) maintainers published a roadmap on August 22, 2025, outlining five priority areas for developers building agents and MCP servers, following the stateless spec released…
Keycloak 26.7, released this month, adds production-ready OAuth 2.1 authorization for AI agents using the Model Context Protocol, enabling per-tool scopes and eliminating the default all-or-nothing ac…
A new IETF draft, Credential Broker 4 Agents (CB4A) by Kenneth G. Hartman, published March 2026, proposes a credential brokering pattern that keeps AI agents from ever holding reusable secrets, instea…
An enterprise cloud and AI architect warns that delegation escalation, not prompt injection, is the biggest architectural risk in enterprise AI today. To secure autonomous multi-agent chains, platform…
AI agents that hold long-lived bearer credentials in their environment, config, or context window create a systemic leak risk because any entity that obtains the credential can reuse it, and agents ma…
Anthropic's Zero Trust for AI Agents framework correctly identifies that agents require stronger security than chatbots, but its recommendation of short-lived bearer tokens fails its own test of makin…