What is MCP governance?

MCP (Model Context Protocol) governance is the structured decision-making around whether an MCP server may connect to your AI agents, at what classification tier, and under what controls. The OWASP MCP Governance & Risk Project frames it as three layers — policy, controls and checklists — answering one question: should this server be allowed, and under what conditions?

What are the four non-negotiable MCP governance rules?

From the OWASP framework: no owner = no approval (every server needs a named owner); no logging = no production use (audit trails are mandatory in production); no scope definition = no access (the data and actions a server can take must be documented); and no review = no enterprise deployment (periodic risk-tier reviews are required). This tool treats each as a hard gate.

What are the MCP risk classification tiers?

The framework uses five tiers: Tier 0 (public data, read-only), Tier 1 (internal non-sensitive read), Tier 2 (sensitive read), Tier 3 (write-capable), and Tier 4 (privileged/critical). Higher tiers demand stronger controls and human approval, and the risk bar this tool applies rises with the tier.

What is the biggest MCP governance risk?

The framework names tool chaining as the primary risk: an MCP server that can invoke other tools or trigger downstream workflows can turn a single approved action into a chain the user never saw. Combined with write access and weak logging, that is where a governed MCP program most often fails.

Does this replace the OWASP MCP framework?

No. It is a fast triage that applies the framework's gates, tiers and risk factors so you can make and record a decision quickly. For the full guidance — inventory, classification, risk scoring and a 90-day rollout — read the OWASP MCP Governance & Risk Project itself.