MCP Find
Browse Servers
Blog
Docs
Submit Server
Back to Directory
/
Security & Auth
io.github.denial-web/agent-immune
AI agent security: prompt injection detection, semantic memory, output scanning, prompt hardening
Security & Auth
v0.2.2
View on GitHub
Learn More
MCP Server Authentication: API Keys vs OAuth 2.1 Explained
The MCP specification adopted OAuth 2.1 in its 2025 revision. Static API keys still dominate for single-developer tools. This guide breaks down when each applies, what PKCE requires, and how the 221 security-category servers on MCPFind reflect real-world adoption patterns.
Read article
MCP Security Basics: What to Know Before Adding Servers
Every MCP server you add to Claude gets access to tools that can read files, call APIs, or query databases. Before connecting one, you should understand the permission model, what prompt injection looks like in practice, and how to evaluate whether a server is trustworthy. This guide covers the basics without requiring a security background.
Read article