Blog
Essays on agent security, MCP infrastructure, and what we learn shipping the gateway to production.
Articles
More posts
Okta is right that non-human identities matter for agentic AI. But identity is the floor of the problem, not the ceiling — and the ceiling is where most of the actual risk lives.
Identity, scope, payload constraints, replayable audit, policy-as-code, and adoption observability — distilled into something you can hand a platform engineer Monday morning.
Page 9 of Anthropic’s Zero Trust for AI Agents quietly invalidates how most agent integrations work today. Here’s what the spec actually says, why most teams won’t build it, and where that gap has to live.
MCP is connecting agents to every API on earth. Nobody’s asking what happens when they get it wrong. Policies, human-in-the-loop, and audit — the seatbelts the AI agent ecosystem still doesn’t have.
