Security series
MCP Security
Model Context Protocol is rapidly becoming the default integration layer between AI clients and tools. It also has six structural risks every deployment needs to account for. This series walks through each, with what breaks if you don’t mitigate.
Posts in this series
Each post focuses on one specific risk class.
-
1.
I Classified 8,000+ MCP Servers by Auth Appropriateness. Most Get It Wrong.A static analysis of 8,216 MCP servers across 4 registries. 50.6% have no auth. But the real question is: do they have the right auth for what they do?March 28, 2026
-
2.
4,500 MCP Servers, 7,840 Tools, Zero Input Validation. Here's the Attack Surface.I analyzed the tool definitions of 8,216 MCP servers. 2,432 expose high-risk input parameters — SQL, file paths, shell commands — with no validation constraints.March 29, 2026
-
3.
Can You Prove What Your AI Agent Did? I Checked 8,216 MCP Servers.74.9% of MCP servers have no mention of audit logging. The MCP spec defines zero audit primitives. Here's why that's a compliance problem.March 30, 2026
-
4.
MCP Servers Proxy Paid APIs With No Rate Limits. I Calculated the Blast Radius.180 MCP servers proxy calls to paid APIs like OpenAI and Stripe. 85% document no rate limits. An agent retry loop can cost $1,080/hour.March 31, 2026
-
5.
MCP Gateway Comparison: ACP vs Composio vs Self-HostedAn honest comparison of MCP gateway options. When to use ACP Cloud, Composio, or build your own with open-source tools.April 1, 2026
-
6.
4 Security Vulnerabilities Hiding in Your MCP Server's Tool SchemaReal CVEs trace back to tool schemas with no constraints. We analyzed 8,216 MCP servers and show the exact JSON patterns that create path traversal, SSRF, injection, and destructive operation vulnerabilities — and how to fix each one.April 2, 2026