Skip to main content

Requirements

Before you begin, you must have the following:
  • A Claude plan. To see if you have a plan, sign in to Claude and then on the sidebar, click your user icon. Sign up for Claude.
    For Claude Desktop, the Transform MCP server does not support users on the Claude Free plan.
  • Claude Desktop installed on your local development machine. Install Claude Desktop.

Install the Transform MCP server

You should be able to instruct Claude Desktop to install the Transform MCP server by asking Claude Desktop the following:
If this does not work, to install the Transform MCP server manually, complete the following steps.If you are able to successfully connect, skip ahead to parse your source files.
Complete one of the following sets of instructions, depending on your Claude plan:

Pro and Max plan users

  1. Start Claude Desktop.
  2. On the sidebar, click your user icon, and then click Settings.
  3. On the side pane, under Settings, click Capabilities.
  4. In the Code execution and file creation section, switch on the Code execution and file creation toggle.
  5. Switch on the Allow network egress toggle.
  6. In the Additional allowed domains list, enter the Transform MCP server’s domain as follows:
  7. Close the Settings dialog.
  8. On the sidebar, click Customize.
  9. On the side pane, click Connectors.
  10. Click Add, then from the drop-down list select Browse connectors.
  11. In the Search connectors box, search for and select the Unstructured Transform connector.
    1. Close the Directory dialog.
    2. Next to Connectors, click the plus icon, and then click Add custom connector.
    3. In the Add custom connector dialog, for Name, enter Unstructured Transform.
    4. For Remote MCP server URL, enter the following:
      https://mcp.transform.unstructured.io
    5. Click Add.
  12. Click Connect, and follow the on-screen directions to finish authenticating. After you finish authenticating, the Transform MCP server is ready for you to start using. Next step: parse your source files.

Team and Enterprise plan users

To allow Claude Desktop to download files to your local directories from Transform MCP, one of your plan’s owners must do the following procedure from their Claude Desktop installation. Grant Claude Desktop access to download files from Transform MCP
  1. Start Claude Desktop.
  2. On the sidebar, click your user icon, and then click Settings.
  3. On the side pane, under Settings, click Capabilities.
  4. In the Code execution and file creation section, switch on the Code execution and file creation toggle.
  5. Switch on the Allow network egress toggle.
  6. In the Additional allowed domains list, enter the Transform MCP server’s domain as follows:
  7. Close the Settings dialog, and then close the Connectors dialog.
Connect a user account to Transform MCP To connect Claude Desktop to Transform MCP from your Team or Enterprise account:
  1. Start Claude Desktop.
  2. On the sidebar, click Customize.
  3. Click Connectors.
  4. Click the Unstructured Transform connector that one of your plan’s owners added.
  5. Click Connect, and follow the on-screen directions to finish authenticating. After you finish authenticating, the Transform MCP server is ready for you to start using.

Parse your source files

Parsing requests have the following limits:
  • Each file must be of a supported file type.
  • Each file must be 50 MB or less in size.
  • Each request must have 10 files or fewer.
  • Only 5 requests can be running at a time.
The Transform MCP server is designed to notify Claude Desktop about these limits. Because of this, Claude Desktop should notify you whenever it encounters a file that exceeds 50 MB in size. Also, Claude Desktop should formulate strategies to send requests that are 10 files or fewer and not cause more than 5 requests to be running at a time. After you connect Claude Desktop to the Transform MCP server, complete one of the following sets of instructions, depending on whether you’re using Claude Desktop in Chat mode or Cowork mode.

Chat mode

  1. In Claude Desktop, on the sidebar, click the speech balloons icon (Chat).
  2. On the sidebar, click New chat.
  3. Add the files you want Transform to parse to your chat session. To do this, next to the Write a message prompt, click the plus icon (Add files, connectors, and more).
  4. Click Add files or photos.
  5. Follow the on-screen directions to finish adding the files.
  6. From your chat message prompt, ask the following:
    Transform parses your input files and delivers its results to you as a set of output files, one output file per input file.
  7. Skip ahead to the next steps.

Cowork mode

  1. In Claude Desktop, on the sidebar, click the to-do list icon (Cowork).
  2. On the sidebar, click Projects.
  3. Click New project.
  4. Click Use an existing folder.
  5. Follow the on-screen instructions to choose the folder that contains the files you want Transform to parse.
  6. From your chat message prompt, ask the following:
    As needed, adapt the phrase in this project in the preceding prompt to indicate a more specific project folder. Also, as needed, guide Claude Desktop as to where to write Transform’s output into the project. Transform parses your input files and delivers its results to you as a set of output files, one output file per input file.

Extract structured data

If you want to pull specific fields from your files, instead of converting the files in full, ask for an extraction. To determine what structured data Transform can extract from your files, use a prompt similar to the following:
Your tool parses each file first, then extracts the field values from the parsed element JSON; this is because the Transform MCP extraction tools read a file’s parsed output, rather than the raw file itself. Transform returns results as JSON matching the schema, wrapped with the source filename so you can tell a batch apart. If you know the specific fields you want, include them in your prompt to guide Transform in extracting just the information that’s important to you. For more information, including prompting patterns and performance considerations, see Structured data extraction.

Next steps

Questions? Need help?