> For the complete documentation index, see [llms.txt](https://docs.darcyiq.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.darcyiq.com/dispatch/welcome-to-dispatch.md).

# Welcome to Dispatch

Managed model access for your organization — one key per person, one bill, and the controls to keep both in check

Every team now wants to use AI models from their own tools — coding agents, editors, scripts, internal apps. Left alone, that turns into a pile of personal provider accounts, keys pasted into chat, and a bill nobody can explain until it arrives.

**Dispatch gives your organization one managed front door to the models.** Each person gets their own API key that works in the tools they already use. Administrators decide what everyone may spend, which models they may call, what content is allowed, and where requests are processed — and everyone can see what it all costs, as it happens.

{% hint style="success" %}
**From key to cost report in minutes**: An admin sets a monthly budget and a per-person allowance. A developer creates a key, follows the Claude Code cookbook, and is coding against Dispatch two minutes later. At the end of the week the admin opens Cost Reporting, groups spend by team tag, and sees exactly which project drove the increase.
{% endhint %}

## Who Is This For?

* **Developers and power users** — one key that works in Claude Code, Cursor, Codex, OpenCode and anything else that accepts an OpenAI-compatible base URL
* **Engineering and IT leaders** — give a whole team model access without handing out provider accounts
* **Finance and operations** — a single bill, budgets that are enforced rather than advisory, and reports you can share
* **Security and compliance** — guardrails on what may be sent, control over which models are reachable, and data residency for where requests are processed

## Key Benefits

| Benefit                            | Description                                                                                      |
| ---------------------------------- | ------------------------------------------------------------------------------------------------ |
| **Works With Your Tools**          | Speaks the OpenAI and Anthropic APIs, with step-by-step cookbooks for popular agents and editors |
| **Personal Keys, Shared Controls** | Everyone manages their own keys; limits and policy belong to the person, not the key             |
| **Enforced Budgets**               | Organization budgets and per-person allowances are hard limits, not warnings                     |
| **Guardrails**                     | Named policies that block sensitive content, cap spend and narrow which models people can call   |
| **Model Access**                   | Choose which model tiers your organization may call                                              |
| **Data Residency**                 | Restrict where requests may be processed                                                         |
| **Cost Reporting**                 | Group, filter and share spend by person, tier or tag                                             |
| **Insights**                       | Checks that flag what's worth fixing, and where Token Optimization is saving money               |

## How Dispatch Works

```mermaid
graph LR
    A["Your tools<br/>(agents, editors, apps)"] -->|"Dispatch API key"| B["Dispatch"]
    B -->|"Budget, guardrails,<br/>model access, residency"| C["AI models"]
    B --> D["Usage &<br/>cost reporting"]
```

1. **Connect** — Point a tool at Dispatch with your personal API key.
2. **Govern** — Requests are held to your allowance, the organization's budget, your guardrail, the models the organization allows and where it may be processed.
3. **Report** — Usage and cost are recorded as requests happen, ready to report on by person, tier or tag.

## Getting Started

{% stepper %}
{% step %}
**Open Dispatch** Launch Dispatch from the DarcyIQ Platform launcher. See [Opening Dispatch](/dispatch/get-started/opening-dispatch.md).
{% endstep %}

{% step %}
**Create a key** Create your personal API key on the API keys page. See [API Keys](/dispatch/use-dispatch/api-keys.md).
{% endstep %}

{% step %}
**Connect a tool** Follow the cookbook for your agent or editor. See [Quickstart](/dispatch/get-started/quickstart.md) and [Cookbooks](/dispatch/use-dispatch/cookbooks.md).
{% endstep %}

{% step %}
**Set the guardrails (admins)** Set a budget, choose which models are allowed, and invite your team. See [Organization Settings](/dispatch/manage/manage.md).
{% endstep %}
{% endstepper %}

## Next Steps

| Goal                                | Documentation                                                   |
| ----------------------------------- | --------------------------------------------------------------- |
| Understand what each role can do    | [Roles & Access](/dispatch/get-started/roles-and-access.md)     |
| Make your first request             | [Quickstart](/dispatch/get-started/quickstart.md)               |
| See what models you can call        | [Models](/dispatch/use-dispatch/models.md)                      |
| Report on spend                     | [Cost Reporting](/dispatch/cost-and-insights/cost-reporting.md) |
| Set budgets and policy for everyone | [Organization Settings](/dispatch/manage/manage.md)             |
| Looking for DarcyIQ itself?         | [DarcyIQ documentation](https://docs.darcyiq.com/)              |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.darcyiq.com/dispatch/welcome-to-dispatch.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.