Barndoor AI

Barndoor AI is an AI gateway and governance platform for controlling access to large language models (LLMs), AI agents, Model Context Protocol (MCP) servers, and MCP tools. The platform routes AI requests, applies policies, and records activity across connected AI systems.

• LLM Gateway: Routes model requests through a central endpoint and can apply provider failover.
• MCP Governance: Manages MCP-server connections and controls access to MCP tools.
• Policy Enforcement: Applies policies by identity, system, MCP server, tool, model, or data type.
• Access Controls: Supports least-privilege access for AI agents and connected tools.
• Audit Logging: Records agent actions, tool calls, and access events.
• Sensitive-Data Controls: Inspects prompts and responses for personally identifiable information, secrets, and other sensitive content before it is sent to external models.
• Usage Controls: Supports budgets, rate limits, and token caps by user, team, or application.
• Observability: Tracks latency, errors, token usage, and cost by model, team, application, or user.
• Deployment Options: Is available as SaaS, private-cloud, and on-premises deployments.

• Audience: Security, IT, platform-engineering, compliance, and AI-engineering teams.
• Use Case: Governing enterprise AI access, controlling agent-to-tool permissions, protecting sensitive data in AI requests, and monitoring AI usage and cost.

• Managed traffic: LLM requests, AI-agent activity, MCP-server connections, and MCP-tool calls
• Control types: Access policies, rate limits, token caps, budgets, audit logs, and a vendor-stated global disable control
• Deployment: SaaS, private cloud, or on premises