Non-Human Identity Management for MCP Agents
Agent actions need tied identities, not borrowed credentials.
Colin Reyes
LetterMCP — Page 4
Agentic AI Foundations
Agent Setup From Slack in Enterprise Environments
Slack agents inherit enterprise system access with no built-in approval controls.
MCP Compliance Requirements for Financial Services Firms
Banks deploying AI agents now must map existing compliance rules to MCP's novel security risks.
Colin Reyes
More From LetterMCP
security platforms that monitor agent context drift during live MCP sessions
New monitoring platforms fill a security gap that existed tools cannot see.
Ayana Brennan
MCP Tool Call Authorization Models Compared
enterprises overlook authorization because the protocol leaves security entirely to implementers.
Reuben Weldon
RBAC Design for MCP Tool Access in Regulated Enterprises
Enterprises must build RBAC into MCP deployments to meet compliance requirements.
Colin Reyes
Unauthorized MCP Tool Calls and Audit Trail Requirements
MCP's trust boundaries collapse in production, enabling unauthorized tool calls at scale.
Reuben Weldon
MCP gateway and catalog platforms compared
Gateways enforce security that MCP left unbuilt; platforms add catalog and identity to the equation.
Mira Okoro
Claude Constitutional AI Limits Under Adversarial Prompting
Adversarial prompting exploits blind spots in principle-based refusal at scale.
Colin Reyes
What MCP Tool Annotations Tell You About Agent Behavior
Four boolean hints preview tool behavior, but only if you trust the server supplying them.
Reuben Weldon
Role of Platform Engineering in AI Agent Enablement
Platform engineering must evolve to govern AI agents with the same rigor it uses for humans.
Mira Okoro
Enterprise AI Agent Deployment Patterns
Enterprises are deploying AI agents faster than security frameworks can govern them.
Mira Okoro