Resonance Proxy
Agent output governance for enterprise AI
“Your AI agents make thousands of decisions. How many did you actually approve?”
AI agents are making procurement decisions, approving invoices, and modifying employee records — at machine speed, 24/7. Your team can't review thousands of agent outputs per day. I'm building a transparent governance layer that intercepts every decision, classifies it by risk, and escalates the ones that matter — before they execute.
The Problem
What needed solving
Enterprise AI agents are making procurement decisions, approving invoices, reallocating budgets, and modifying employee records — at machine speed, 24/7. Your team can't review thousands of agent outputs per day. But they also can't afford to let agents run unchecked.
The result? Companies either slow everything down with manual approval gates on every action (defeating the purpose of agents), let everything through and discover problems after the damage is done, or build custom governance logic into every agent, duplicating effort across teams. None of these scale.
The Solution
How I approached it
Resonance Proxy is a transparent governance layer that sits on the Model Context Protocol (MCP) — the open standard for AI agent communication. It intercepts agent actions at the protocol level, so you don't need to modify your agents.
Every action gets classified by risk. Low-risk decisions auto-approve and log. Medium-risk items batch into human-digestible digests for periodic review. High-risk actions escalate immediately — before they execute. The result is governance that scales with your agent fleet instead of against it.
How It Works
Under the hood
Resonance Proxy sits between your AI agents and the actions they take as an MCP middleware layer. When an agent issues a tool call over SSE, the proxy intercepts it before execution. A classification engine evaluates the action against configurable risk policies — amount thresholds, affected systems, data sensitivity, and custom business rules.
Based on classification, the action routes to one of three paths: auto-approve with audit logging, batch for periodic human review, or escalate for immediate approval. A governance dashboard gives teams a single pane to review, approve, and audit all agent activity. The entire system is protocol-native — no agent modifications required, no SDK to integrate, no vendor lock-in.
Impact
Results and outcomes
Resonance Proxy targets the emerging enterprise need for AI agent governance — a market projected to grow as 40% of enterprise apps embed AI agents by end of 2026. By operating at the protocol level (MCP), it provides a universal governance layer that works across any MCP-compatible agent without modification.
The batched review model means a team of five can govern a fleet of hundreds of agents, reviewing only what matters while maintaining a complete audit trail of every automated decision.