AI governance
AI Governance Best Practices for Agentic Tool Access
Organizations must inventory all deployed agents before any other security controls can work.
Greta Wexler
· 9 min read
MCP Report — Page 2
MCP Gateway Platforms Compared on Prompt Interception and Sanitization Capabilities
Early interception stops tool-poisoning attacks; late filtering just documents the damage.
Greta Wexler
MCP Tool Poisoning Attacks Explained
Attackers hide malicious instructions in tool metadata that AI agents trust automatically.
Greta Wexler
MCP Architecture Components and Data Flow
How MCP collapses the N×M integration problem into N+M through standardized roles.
Greta Wexler
MCP Server Security in Multi-Tenant Environments
Shared MCP infrastructure requires tenant isolation layers the protocol doesn't enforce by design.
Isaac Dieng
MCP Gateway and Server Management Platforms Compared
Most "MCP gateway" products are not built for production scale.
Isaac Dieng
MCP Rug Pull Risks in Third-Party Servers
Most MCP servers change their behavior after approval with no re-review required.
Adeel Choudhury
MCP Rug Pull Protection Mechanisms Across Leading Tools
Rug pulls exploit the MCP protocol's lack of cryptographic signing and re-approval workflows.
Lucia Sato