🤖 AI Gateway
Control Core AI Gateway provides enterprise authorization and runtime controls for:
- External and internal LLM routes
- MCP server and tool interactions
- AI agent and application traffic
- Guardrail enforcement and policy-backed decisions
- Forensics-grade traceability for investigations
📌 Why this matters
AI rollouts fail when teams must modify each application to add controls. With Control Core, AI traffic is enforced at the Bouncer layer so protected resources can stay unchanged.
🏗️ Core operator workflow
- Open Settings -> AI Pilot.
- Select the bouncer that fronts the target application or AI endpoint.
- Configure token and cost controls, MCP/LLM guardrails, and content-safety actions.
- Validate decisions and traces in the built-in forensics/intelligence views.
- Promote controls from Sandbox to Production after validation.
📌 Environment model
Control Core runs as one control plane instance with two isolated environments:
- Sandbox for development and validation
- Production for live enforcement
Isolation applies to policies, bouncers, resources, AI connections, PIP connections, and action destinations.
The only cross-environment flow is controlled promotion from sandbox to production.