Latest
ResearchAn Open Letter to Dario AmodeiResearchAnnouncing State of Enterprise Agent Permissions 2026ResearchClaude Code Security Benchmark: What It Can Access on a Typical Developer LaptopReportState of Enterprise Agent Permissions 2026Research10 Actions Your EDR May Allow an AI Coding Agent to PerformResearchCISO Guide to Agent Least PrivilegeResearchAI Agent Exposure Assessment ChecklistResearchClaude Code vs Human Permissions: What Should Be Different?
All topics

Enterprise AI agent security

MCP access control

Access control for MCP is not a single gate at the network edge. It is per-actor, per-operation permission on the system where the tool executes — the same model as shell and API access.

MCP control point

Tool invocation expands authority in one call — evaluate MCP bridges to shell, files, and cloud on the host.

Tool-using agent

MCP client

MCP server

Tool surface

Local shell

Block or audit

Secret paths

Least privilege

Prove exposure in your environment

The AI Agent Exposure Assessment runs Audit Mode on a pilot developer group for fourteen days and delivers evidence your security leadership can act on — then you enforce what matters.

Get your exposure report

Implement access control

  • •Map MCP tools to normalized actions (read, write, exec, network).
  • •Apply least privilege per developer or agent class.
  • •Prove effective coverage with fourteen days of Audit Mode telemetry.

Related

MCP use case
Get your exposure report AgentProtect