> For the complete documentation index, see [llms.txt](https://docs.uptiq.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph/upload-and-manage-a-govern-graph.md).

# Upload and manage a govern graph

Upload a graph package, add and restore versions, rename a graph, edit a node, and attach the graph to an agent.

Load an institution's knowledge graph into Qore, keep it current with new versions, and make it available to agents. Each task below stands on its own.

## Before you begin

* Have a role that can manage govern graphs: Governance Manager, Account Admin or Org Admin. These roles have the **Ingest source & extract knowledge graph candidates** capability. See [Roles](https://docs.uptiq.ai/ai-workbench-qore/control-center/admin/roles.md).
* Prepare a graph package: one `.json` file, up to 80 MB, with a top-level `nodes` object keyed by category and a top-level `edges` array. Give every node an `id`. See [Graph package](https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph.md#graph-package).
* To attach a graph to an agent, have the Developer, Account Admin or Org Admin role.

## Upload a graph

1. Open **Control Center → Govern → Govern Graph**.
2. Select **Upload Govern Graph**.
3. Enter a **Graph name**, for example `CCU Govern Graph`.
4. Add the **Graph package file**.
5. Select **Upload**.

The package is validated, then written into the graph. While it builds, the row shows **Processing** and the current stage: Queued, Validating the package, Writing nodes, then Writing relationships. The row ends as **Completed** or **Failed**. If it fails, the **Failed** badge gives the reason. See [Troubleshoot govern graphs](https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph/troubleshoot-govern-graphs.md).

## Build a graph in your own Neo4j

Use this option in a development environment only.

1. In the upload dialog, select **Use your own Neo4j (Dev Only)**.
2. Enter the **Connection URL**, **Username** and **Password**. Enter a **Database**, or leave it empty to use the server's default database.
3. Select **Test connection** to check the details before the build.
4. Select **Save**.
5. Select **Upload**.

The graph is built in your server instead of the platform's. Saving doesn't connect, so a wrong address, password or database appears as an error when the graph is built. Have the user ready with write access and permission to create constraints and indexes. On Neo4j Aura, the address and credentials are in the instance details and the credentials file. For the field details, see [Own Neo4j settings](https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph.md#own-neo4j-settings).

## Retry a failed build

1. Open the `⋮` menu on the graph's row.
2. Select **Retry**.

**Retry** shows only when a retry is possible. The build uses the package already uploaded. A failed build retries on its own first, and the row says when automatic retries are used up.

## Upload a new version

1. Open the `⋮` menu on the graph's row.
2. Select **Upload new version**.
3. Drag a `.json` file onto the dashed area, or select the area to choose one.
4. Select **Upload**.

The new package becomes the next version of the same graph. It's validated and built the same way as a first upload. Agents keep reading the current version until the new one finishes building, and then switch. If the build fails, nothing changes for agents.

## Restore an earlier version

1. Open the `⋮` menu on the graph's row.
2. Select **Version history**.
3. Find the version to restore. Select **View graph** to look at it first.
4. Select **Restore**.
5. Confirm in the **Restore version N?** dialog.

Agents read the restored version right away. Nothing is rebuilt, and the later versions are kept. The restore is recorded as a new version.

## Rename a graph

1. Open the `⋮` menu on the graph's row.
2. Select **Edit**.
3. Change the **Graph Name**.
4. Select **Save**.

Renaming changes only the name. It doesn't change the package, the nodes or the versions.

## Edit a node

1. Open the graph, and then open **Graph Explorer**.
2. Select the node. Use **Search nodes** to find it by name.
3. Select **Edit node**, the pencil in the panel.
4. Change the properties. **Name** is required, **Features** takes one item per line, and **Confidence Score** takes a number. The ID can't be changed.
5. Select **Save**, or **Cancel** to discard your changes.

For which values you can edit, see [Node editing](https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph.md#node-editing).

## Ask a graph a question

1. Open the graph, and then open **Ask the Agent**.
2. Choose a version in the select.
3. Type a question, or select one of the starter questions and send it.

The answer is what an agent would give, grounded in the graph. A **Graph Query** chip appears when the platform ran a graph query to produce it.

## Regenerate or edit the graph guides

1. Open the graph, and then open **About Graph**.
2. Select **Graph Summary** or **Traversal Guidelines**.
3. Select **Generate** when the tab has no text yet. Otherwise change the text in the editor, or select **Regenerate** to have the platform write a new one.
4. Select **Save** to keep your version.

Each save or regeneration is stored as a new version of the guide for that graph version. Agents use the guides when they query the graph.

## Attach a graph to an agent

1. In Agent Builder, select **+ Add Knowledge** on the canvas.
2. In **Use Existing Knowledge**, select the graph.
3. Select **Add**.

The agent queries the graph with Knowledge Graph Search. You can't create or attach a graph by attaching the package JSON in the Agent Builder chat. Upload it first. See [Attaching a govern graph](https://docs.uptiq.ai/ai-workbench-qore/console/agent-builder/build/attaching-knowledge-to-an-agent.md#attaching-a-govern-graph).

{% hint style="warning" %}
Agents read unconfirmed nodes. Review what an agent quotes from the graph before you rely on it with customers.
{% endhint %}

## What's next

<table data-view="cards"><thead><tr><th>Title</th><th>Description</th><th data-card-target data-type="content-ref">Link</th></tr></thead><tbody><tr><td><strong>Govern Graph</strong></td><td>Reference for nodes, relationships, tabs and agent skills.</td><td><a href="https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph.md">Govern Graph</a></td></tr><tr><td><strong>Troubleshoot govern graphs</strong></td><td>Fix failed uploads and unexpected graph content.</td><td><a href="https://docs.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph/troubleshoot-govern-graphs.md">Troubleshoot govern graphs</a></td></tr></tbody></table>


---

# 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.uptiq.ai/ai-workbench-qore/control-center/govern/govern-graph/upload-and-manage-a-govern-graph.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.
