> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.vellum.ai/product/prompts/collaboration/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.vellum.ai/_mcp/server.

# Collaborate on Prompts with Vellum Prompt Sandbox

> Explore Vellum Prompt Sandbox for rapid iteration, collaboration, and sharing of model-generated prompts. Save, tag, and track progress easily.

The Vellum Prompt Sandbox is a powerful tool for rapid iteration and
collaboration between multiple models and prompts. Save, tag,
and share your work with ease using the features outlined below.

### History

Every model-generated response and respective prompt are saved
as history items, giving you access to a detailed record of your
work. To access history items, simply activate the toggle button
located at the top right side of the Sandbox, and all history items
will appear on the right side of your screen.

![Sandbox History](https://storage.googleapis.com/vellum-public/help-docs/prompt-history.png)

### Tracking Progress, Collaborating, and Tagging

The Prompt Sandbox is designed to help you iterate on prompts and model
providers until you find the perfect fit for your needs. With the
history feature, you can keep track of your team's work in an organized
way by only keeping the iterations you choose to, through the `save` button.

Everyone working on the same sandbox can see each other's history items,
and you can also tag them to keep better track of your work.

![Tagging History](https://storage.googleapis.com/vellum-public/help-docs/tagging-history.png)

### Share Your Work

At any point in time, you can easily share your work with anyone in
your organization through a URL by using the “invite” button located
at the bottom left side of the page.

![Inviting Teammates](https://storage.googleapis.com/vellum-public/help-docs/inviting-teammates.png)

> Explore Vellum Prompt Sandbox for rapid iteration, collaboration, and sharing of model-generated prompts. Save, tag, and track progress easily.