Protocols and standards
Available now
Halo and Model Context Protocol (MCP)
Govern MCP tool traffic through Halo’s dedicated MCP and tool gateway.
How Halo works with Model Context Protocol (MCP)
Model Context Protocol is an open protocol introduced by Anthropic in November 2024 for connecting AI applications to tools and data.
Halo’s MCP and tool gateway applies identity, policy, approvals, budgets, and evidence to supported agent tool traffic. Available tools and actions depend on the connected server and its configuration.
Governance
What Halo governs for Model Context Protocol (MCP)
Agent identity
Keep agent identity separate from user identity, credentials, tools, and action scope.
Tool-level policy
Evaluate tool requests against configured policy in the MCP and tool gateway.
Higher-impact actions
Require human approval for actions configured as higher impact.
Credentials and evidence
Broker credentials at the gateway and retain reviewable records of decisions and approvals.
Setup
Get connected in three steps
- 1
Plan the MCP path
Identify the agents and MCP servers that should route through the Halo tool gateway.
- 2
Set identity and policy
Configure scoped access, tool-level policy, and approval gates for higher-impact actions.
- 3
Test and review
Check gateway decisions and evidence, then adjust policy before expanding use.
FAQ
Questions, answered
MCP is an open protocol for connecting AI applications to tools and data.
Halo’s MCP and tool gateway applies configured identity, policy, approvals, budgets, and evidence controls to agent tool traffic.
No. Halo governs supported traffic through its gateway and does not certify connected servers.
Related capabilities
More protocols and standards
Put a halo around the AI your business runs.
See discovery, runtime policy, approvals, and evidence working together on your own use cases.