# Cortega

> AI Border Gateway, Endpoint Guard, and Cortega Agents: governance for every AI request, on the gateway and on the device.

See /for-ai-agents for how an AI agent should use this site.

## Pages

- [Cortega](/index.md): Cortega AI Governance Platform. AI Border Gateway, Endpoint Guard, and Cortega Agents.
- [For AI Agents](/for-ai-agents.md): How an AI assistant (Claude, ChatGPT, Gemini, or any other) can read this site accurately.
- [AI Border Gateway](/docs/products/ai-border-gateway.md): Cortega's core traffic-governance product. Every LLM and MCP tool call is identified, routed, and checked before it reaches a model or a tool.
- [AI Border Gateway capabilities](/docs/products/ai-border-gateway-capabilities.md): The full capability reference for AI Border Gateway, authentication, authorization, routing, guardrails, cost controls, and observability.
- [Endpoint Guard](/docs/products/endpoint-guard.md): Local interception and guardrails for AI traffic that never goes through your gateway. Desktop apps, browsers, and local MCP tools.
- [AI Verifier](/docs/products/ai-verifier.md): Cortega's citation and hallucination checking agent, plus a desktop app that brings the check to the person who got the answer.
- [Platform](/docs/platform.md): The shared foundation every Cortega component runs on. Identity, observability, audit logging, RBAC, benchmarking, and the API, in one place.
- [Cortega Agents](/docs/platform/cortega-agents.md): What a Cortega Agent is, the full catalog, and which ones are mature enough to rely on today.
- [The AI governance problem](/docs/get-started/the-ai-governance-problem.md): The recurring problems that show up once more than one person or agent is calling an LLM or an MCP tool, and which Cortega product addresses each.
- [Choosing the right product](/docs/get-started/choosing-a-product.md): AI Border Gateway and Endpoint Guard are independent capture points. Cortega Agents read from either one.
- [Online vs Hosted](/docs/get-started/online-vs-hosted.md): On-Prem, SaaS, or Dedicated. The signals that point toward each, and what changes for you at each choice.
- [Model Intelligence](/docs/tools/model-intelligence.md): A curated, weekly-refreshed catalog of LLM quality, cost, and speed data. Free to use standalone, and available as a plugin on AI Border Gateway.
- [Cortega Assist](/docs/tools/cortega-assist.md): A free, hosted service that answers questions about Cortega and gateway configuration for any connected agent, no account required.
- [Workload Analyzer](/docs/plugins/workload-analyzer.md): Classifies your AI traffic into workload categories, then uses that classification with Model Intelligence to recommend which model to run.
- [Install overview](/docs/install.md): Three downloads. The Platform (same installer as AI Border Gateway), Endpoint Guard, and AI Verifier, plus the hosted option if you'd rather not install anything.
- [Try it on your laptop](/docs/install/local-quickstart.md): Run Cortega locally with Docker Compose to try it before committing to a real install.
- [AWS, scripted](/docs/install/aws.md): Cortega provisions EC2 and deploys onto it, single-node or multi-node, from one script.
- [Terraform](/docs/install/terraform.md): One shared deployer for AWS, Azure, or GCP.
- [Existing environment](/docs/install/existing-environment.md): Run Cortega on Docker hosts you already manage, with no provisioning step.
- [Guides](/docs/guides.md): Day-2 usage guides for running Cortega. Published here in full, and shipped inside every install package too.
- [User guide](/docs/guides/user-guide.md): Step-by-step guide for platform administrators to configure Cortega from zero to governed LLM traffic.
- [User management guide](/docs/guides/user-management.md): How to manage accounts, SSO, roles, permissions, and access control.
- [Identity guide](/docs/guides/identity.md): How to configure caller identities for LLM and MCP traffic.
- [Guardrails guide](/docs/guides/guardrails.md): How to configure LLM and MCP request/response guardrails.
- [MCP servers guide](/docs/guides/mcp-servers.md): How to register upstream MCP servers and discover their tools.
- [MCP authorization guide](/docs/guides/mcp-authorization.md): How to grant teams access to MCP servers and tools.
- [Pass-through routing](/docs/guides/pass-through-routing.md): How to let clients use their own upstream provider or MCP-server credentials while Cortega still governs the traffic.
- [Client identification guide](/docs/guides/client-identification.md): How Cortega records which application made each governed request.
- [Endpoint Guard deployment guide](/docs/guides/edge-agent-deployment.md): How to try Cortega Endpoint Guard on one machine, then enroll licensed devices with MDM or administrator-provided certificates.