Secure Passage Adds MCP Support to Truman for AI-Controlled Security Systems
Secure Passage is developing Model Context Protocol server capabilities for Truman, its identity-driven Physical AI Operating System. The capabilities are designed to let compatible AI applications and agents access connected security, infrastructure and autonomous systems through a common interface.
- Secure Passage is developing MCP server capabilities for Truman.
- Compatible AI applications and agents can access connected vendor systems through Truman.
- Supported workflows include investigations, operational awareness and coordinated response.
- Configured permissions and human approval requirements govern available actions.
- The capabilities can extend to legacy systems through existing integrations.
A common interface for connected systems
Organizations often use a mix of cameras, access control platforms, building systems, sensors, robots and drones. According to Secure Passage, these technologies can hold separate parts of the operational picture. Truman brings their inputs together with identity, location and operational context.
Its MCP server capabilities are intended to let compatible AI tools work through Truman rather than establish separate connections to every vendor platform. The company says the approach can support cross-system investigations, operational awareness and coordinated response.
Potential workflows include combining identity, access, video detection and location information to understand an event, asking natural-language questions about facilities and equipment, and initiating supported workflows across connected systems. Robot and drone capabilities can also be incorporated into supported investigation and response workflows. The functionality available in each deployment will depend on its integrations, capabilities and configured permissions.
Extending AI access to existing infrastructure
Secure Passage says Truman can extend MCP-enabled AI capabilities to connected systems that do not natively support the protocol, including legacy access control platforms. Where existing integrations support it, AI agents can request operations such as managing users, updating access permissions and initiating workflows through Truman's common interface.
The company says the architecture can also extend across connected operational technology, cyber, infrastructure and autonomous systems. That would allow organizations to expand the uses of existing investments without replacing them.
“What's significant isn't simply adding MCP support to another security product. It's making that capability available across an entire connected environment, including systems never designed for AI interaction,” Andy Jones, Secure Passage's chief technology officer, said in the announcement.
Identity and policy remain central
Truman's model combines AI interpretation with deterministic controls. Secure Passage says defined policies govern which actions Truman can execute automatically and which require human approval. The MCP capabilities are intended to apply that operating model to compatible AI clients, with access and actions tied to an organization's configured controls.
“An AI agent's ability to request an action should never be confused with permission to execute it,” Gary Fish, chairman and CEO of Secure Passage, said. The company says its aim is to provide operational context and governed access so AI can connect intelligence to action while the organization remains in control.
The development supports Secure Passage's vision for Truman as a vendor-neutral Physical AI Operating System connecting physical security, digital identity, infrastructure monitoring and autonomous systems. For users, the announced capabilities could provide a common path for AI-assisted investigations and responses across existing systems, subject to the integrations, permissions and approval requirements configured in each deployment.
