> For the complete documentation index, see [llms.txt](https://docs.maiagent.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.maiagent.ai/maiagent-user-guide/en/build/overview.md).

# MaiAgent Build Overview

MaiAgent Build is a platform for building your organization's own AI assistants: define roles, knowledge, tools, and permissions, connect databases and enterprise systems, test before launch, and moni

**Turn your ideas into enterprise-grade Agents.**

MaiAgent Build lets IT staff and technical partners build their organization's own AI assistants: define roles, knowledge, tools, and permissions, and connect databases and enterprise systems so AI assistants can retrieve data, call systems, and follow workflows to get things done. The AI assistants used in MaiAgent Work and MaiAgent Serve are also built and managed here.

## Products <a href="#products" id="products"></a>

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Agent Builder</strong></td><td>Define role instructions, knowledge bases, databases, tools, skills, and Agent Teams in one place, and configure an AI assistant ready for deployment.</td><td><a href="/maiagent-user-guide/en/build/agent-builder.md">Agent Builder Overview</a></td></tr><tr><td><strong>AI Gateway</strong></td><td>Connect your organization's own model providers or self-hosted models, manage the model catalog and fallback model groups, and monitor traffic and anomalies.</td><td><a href="/maiagent-user-guide/en/build/ai-gateway.md">AI Gateway</a></td></tr><tr><td><strong>AgentOps and Governance</strong></td><td>Test datasets, automated evaluations, AI assistant monitoring, and tool execution records; Hook checks personal data and content as messages enter and leave the AI.</td><td><a href="/maiagent-user-guide/en/build/agent-ops.md">AgentOps</a></td></tr><tr><td><strong>Developer Tools</strong></td><td>View API call logs, or use MaiAgent MCP to let external AI tools such as Claude and Cursor operate the platform directly.</td><td><a href="/maiagent-user-guide/en/build/maiagent-mcp.md">MaiAgent MCP</a></td></tr></tbody></table>

## Agent Builder Modules <a href="#modules" id="modules"></a>

| Module                                                                                                                                           | Enables AI assistants to…                                                                                                                                                     |
| ------------------------------------------------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [AI Assistants](/maiagent-user-guide/en/build/agent.md)                                                                                          | Understand requests and generate responses based on role instructions; configure [Agent Schedules](/maiagent-user-guide/en/build/agent/schedule.md) to run tasks at set times |
| [Knowledge Bases](/maiagent-user-guide/en/build/knowledge-base.md)                                                                               | Answer questions using enterprise documents, FAQs, and web content                                                                                                            |
| [Databases](/maiagent-user-guide/en/build/database.md)                                                                                           | Query structured data using natural language (Text to SQL)                                                                                                                    |
| [Tools](/maiagent-user-guide/en/build/tools.md)                                                                                                  | Call external systems and perform actions through MCP or APIs                                                                                                                 |
| [Skills](/maiagent-user-guide/en/build/skills.md)                                                                                                | Follow a set of multistep procedures (SOPs) to complete tasks                                                                                                                 |
| [Code Interpreter](/maiagent-user-guide/en/build/agent/code-interpreter.md), [Browser Tool](/maiagent-user-guide/en/build/agent/browser-tool.md) | Run code to process files, or operate a browser to complete web workflows                                                                                                     |
| [Agent Teams](/maiagent-user-guide/en/build/teams.md)                                                                                            | Collaborate across multiple AI assistants with different specialties                                                                                                          |
| [Hook Message Interception](/maiagent-user-guide/en/build/hook.md)                                                                               | Automatically mask personal data or intercept content before it enters the AI or before a response is sent                                                                    |

Not sure which module to use? See [I want to do X. What should I use?](/maiagent-user-guide/en/build/agent-builder/decision-guide.md).

## Build Workflow <a href="#build-flow" id="build-flow"></a>

{% stepper %}
{% step %}

### Create an AI Assistant <a href="#create-agent" id="create-agent"></a>

[Create an AI assistant](/maiagent-user-guide/en/build/agent/setup.md) and write its [role instructions](/maiagent-user-guide/en/build/agent/system-prompt.md); you can also apply an official template from the [Agent Template Library](/maiagent-user-guide/en/build/agent/agent-templates.md) and customize it.
{% endstep %}

{% step %}

### Connect Knowledge and Capabilities <a href="#connect-capabilities" id="connect-capabilities"></a>

Attach a [knowledge base](/maiagent-user-guide/en/build/knowledge-base/km.md), then add [databases](/maiagent-user-guide/en/build/database.md), [tools](/maiagent-user-guide/en/build/tools.md), and [skills](/maiagent-user-guide/en/build/skills.md) when the assistant needs to retrieve data or perform actions.
{% endstep %}

{% step %}

### Test Before Launch <a href="#test-before-launch" id="test-before-launch"></a>

Use [test datasets](/maiagent-user-guide/en/build/agent-ops/test-datasets.md) and [automated evaluations](/maiagent-user-guide/en/build/agent-ops/evaluations.md) to verify response quality, then launch for external users through [MaiAgent Serve](/maiagent-user-guide/en/serve/overview.md) channels, or make the assistant available to employees in [MaiAgent Work](/maiagent-user-guide/en/work/overview.md).
{% endstep %}

{% step %}

### Monitor Continuously After Launch <a href="#monitor-after-launch" id="monitor-after-launch"></a>

In [AgentOps](/maiagent-user-guide/en/build/agent-ops.md), review response quality and [tool execution records](/maiagent-user-guide/en/build/agent-ops/tool-execution-records.md), and [add conversations with poor answers to the FAQ](/maiagent-user-guide/en/build/agent-ops/add-to-faq.md) to keep improving.
{% endstep %}
{% endstepper %}

## Permissions and Governance <a href="#governance" id="governance"></a>

Assign access permissions for AI assistants, knowledge bases, and conversation platforms by role. Organization owners can also restrict which models members and AI assistants can call. See [Role-Based Resource Access](/maiagent-user-guide/en/org/roles/group-resource-access.md), [Model Access Permissions](/maiagent-user-guide/en/org/model-access.md), and [Third-Party Login (SSO)](/maiagent-user-guide/en/org/sso.md). Use the [Activity Log](/maiagent-user-guide/en/org/activity-log.md) to look up and export who changed which settings and when.

## Further Integration <a href="#further-integration" id="further-integration"></a>

To integrate MaiAgent programmatically, see the technical guide and API documentation:

* [API Quickstart](https://docs.maiagent.ai/tech/api-integration/quickstart)
* [AI Gateway: OpenAI-Compatible API Integration](https://docs.maiagent.ai/tech/ai-gateway/ai-gateway-openai-compatible)
* [Self-Hosted MCP Server Integration Specifications](https://docs.maiagent.ai/tech/api-integration/self-hosted-mcp-server)
* [API Documentation](https://docs.maiagent.ai/api)

## Use Cases <a href="#use-cases" id="use-cases"></a>

* [External Data Integration](/maiagent-user-guide/en/build/integrate.md): Connect external API data to AI assistants
* [AI Assistant Optimization Consultant](/maiagent-user-guide/en/build/optimizer-assistant.md): Automatically diagnose response quality and suggest adjustments to role instructions and knowledge bases


---

# 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.maiagent.ai/maiagent-user-guide/en/build/overview.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.