> ## Documentation Index
> Fetch the complete documentation index at: https://docs.corepillar.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# CorePillar documentation

> Guides and reference for ModelHub, CorePillar's governed AI chat client, and AgentHub, the platform for evaluating, tracing and governing AI agents.

CorePillar ships two products that share one account, one organization and one set of model configurations.

<CardGroup cols={2}>
  <Card title="ModelHub" icon="message-square" href="/modelhub/introduction">
    A chat client for employees. Pick a model or let ModelHub route each question to the cheapest model that can answer it, with privacy screening on every message.

    Served at `modelhub.corepillar.ai`.
  </Card>

  <Card title="AgentHub" icon="bot" href="/agenthub/introduction">
    The platform for teams building AI agents. Manage models and keys, ingest traces, run evaluations, estimate cost, and watch org-wide AI usage.

    Served at `pillarforge.corepillar.ai`.
  </Card>
</CardGroup>

## Which one do I need?

| You are... | Use | Start with |
| - | - | - |
| An employee who wants to use AI at work | ModelHub | [ModelHub quickstart](/modelhub/quickstart) |
| An administrator rolling AI chat out to staff | AgentHub, then ModelHub | [Roll out ModelHub](/modelhub/admin/rollout) |
| A developer instrumenting an agent | AgentHub | [Send your first trace](/agenthub/developers/tracing-quickstart) |
| An AI builder comparing models and measuring quality | AgentHub | [AgentHub quickstart](/agenthub/quickstart) |
| Integrating with the REST API | AgentHub | [API reference](/api-reference/introduction) |

## How the two fit together

```mermaid theme={null}
flowchart LR
  A[Admin in AgentHub] -->|adds provider keys and models| M[(Organization models)]
  A -->|approves emails, invites staff| U[Employees]
  U -->|chat| MH[ModelHub]
  MH -->|uses| M
  MH -->|usage and screening records| D[AI Usage dashboard in AgentHub]
  D --> A
```

Administrators configure models in AgentHub. Employees use those models through ModelHub. The spend and privacy activity from ModelHub shows up in AgentHub's [AI Usage](/agenthub/ai-usage) dashboard.

<Note>
  Both products read the same organization data. What someone can see depends on which product URL they open and on their role. See [Access control](/modelhub/admin/access-control).
</Note>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.