> ## Documentation Index > Fetch the complete documentation index at: https://openworklabs.com/docs/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > For agents: install the OpenWork desktop app with `brew install --cask openwork` (macOS) or from https://openworklabs.com/download. Do not run `npx openwork`; the npm package named `openwork` is a different project. > Connect an agent to OpenWork with the MCP Gateway: `claude mcp add --transport http openwork https://api.openworklabs.com/mcp/agent`. Teams sign up at https://app.openworklabs.com?mode=sign-up. > Step-by-step agent guide: https://openworklabs.com/docs/start-here/use-openwork-from-an-ai-agent.md > Connect Slack (any organization admin): https://openworklabs.com/docs/slack/connect-slack.md. OpenWork in Slack (@openwork bot) is private alpha for approved organizations. Setup guide: https://openworklabs.com/docs/slack/overview.md. Agent skill: https://openworklabs.com/.well-known/agent-skills/set-up-openwork-slack/SKILL.md. # Users and teams > See who can use which AI Gateway models, their spend limit and what they spent, and change access for everyone, a team or one person. The **Users & Teams** tab answers one question: who can use which models, how much they can spend, and how much they've spent. You change access on each provider; this tab shows the result for every person and team in one place. ## Where to find it In OpenWork Cloud, open **AI Gateway** and select the **Users & Teams** tab. Seeing it needs three permissions: **View Gateway providers**, **View Gateway usage**, and **View usage limits**. Organization owners and admins have them by default. ## What the tab shows At the top, a row for **Everyone in** your organization shows what every member gets, including people who join later. Below it, switch between **People** and **Teams**, or use **Filter by name**. Each row shows: | Column | What it means | | - | - | | **Can use** | The providers whose models this person or team can use. **Nothing yet** means none. | | **Limit** | Their spend limit and where it comes from, or **No limit**. | | **Last 31 days** | Their estimated spend. | Spend is an estimate from public model prices, not your provider's bill. See [Token costs](https://openworklabs.com/docs/ai-gateway/token-costs.md). ### Open a team Click a team to see only its members. A strip at the top shows the team's providers, limit, and spend, with **Set a team limit** or **Edit team limit**. No teams yet? Click **Create a team**. Teams are managed in **Members**. ### Open a person Click a person to see: * **Last 31 days**: their total estimated spend. * **What they can use**: each provider they can use. Click **Change in AI Providers** to change it. * **Spend limit**: whether they're **Within the limit**, **Over the limit**, or **Paused**, with **Set a limit** or **Edit**. * **Models they use**: spend per model. ## Give people models Access is set on each provider, not on the person. To give someone models: **1. Open the provider** Go to **AI Gateway › AI Providers** and click the provider. To add a new one, click **Add a provider**. **2. Choose who can use it** Under **Who can use it**, turn on **Everyone in the organization**, or turn it off and click **Add person** or **Add team**. **3. Choose which models** Under **Models**, pick **All** models (for example **All Anthropic models**), or **Only the ones I pick** and select them. **4. Save** Click **Save changes**. Members' apps pick up the new models on their next sync. A person can use a provider's models if any route gives them access: everyone, one of their teams, or them directly. [OpenWork Models](https://openworklabs.com/docs/ai-gateway/openwork-models.md) is different: when it's on, everyone in the organization can use it. ## Shared key or each person's own account Each provider has a **Key** setting: * **Shared API key**: you paste one key, and everyone with access uses it. The key stays on OpenWork's servers; members never see it. * **Each member signs in**: each person connects their own account, such as Google sign-in for [Vertex AI](https://openworklabs.com/docs/ai-gateway/google-agent-platform.md), AWS IAM Identity Center for [Amazon Bedrock](https://openworklabs.com/docs/ai-gateway/amazon-bedrock.md), Microsoft Entra ID for [Microsoft Foundry](https://openworklabs.com/docs/ai-gateway/microsoft-foundry.md), or a personal key for [LiteLLM](https://openworklabs.com/docs/ai-gateway/litellm.md). Only some providers support this. ## Stop members from adding their own keys On **AI Gateway › AI Providers**, the **Who can use models** row shows your organization's policy. Click **Change**: * **Only models you provide**: members see only what you set up in AI Gateway and OpenWork Models, and can't add their own provider keys. You can still allow **Admins may still add their own keys on their device**. * **Any model**: members can also add their own provider keys on their device. Your AI Gateway models still show up. The same dialog has **Free starter model (Auto)**, which lets members fall back to OpenWork's free, rate-limited model. Changes apply to every member, on desktop and the web, the next time they open OpenWork. ## Set limits To cap spend for everyone, a team, or a person, see [Limits](https://openworklabs.com/docs/ai-gateway/limits.md). The **Limit** column on this tab shows which limit applies to each row. ## Model groups, credential sets and access grants You may see these terms in permissions, the API, or a member's **Technical details** in My Library. They're the pieces behind the screens above: * A **model group** is a set of models from one provider: all of them, or only the ones you picked. * A **credential set** is how those models are paid for: the shared API key, or each member's own sign-in. * An **access grant** gives a model group to everyone in the organization, a team, or one person. The **Who can use it**, **Models** and **Key** sections on a provider set all three for you. ## Related * [AI Gateway overview](https://openworklabs.com/docs/ai-gateway/overview.md) * [Overview tab: spend and usage](https://openworklabs.com/docs/ai-gateway/usage-dashboard.md) * [Limits](https://openworklabs.com/docs/ai-gateway/limits.md) * [Managing members and RBAC](https://openworklabs.com/docs/cloud/members-and-rbac.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.