UNS-MCP (Unstructured Model Context Protocol) MCP Server

What does “UNS-MCP” MCP Server do?

The UNS-MCP (Unstructured Model Context Protocol) Server is a specialized MCP server implementation designed for seamless interaction with the Unstructured API. It serves as a bridge between AI assistants and external data sources, connectors, and workflows, enabling enhanced automation and integration within development workflows. With UNS-MCP, developers and AI clients can perform tasks such as listing sources and workflows, managing connector lifecycles, and orchestrating data pipelines—all through standardized MCP tools. By exposing workflow and connector management as tools, the UNS-MCP Server empowers developers to automate routine data engineering tasks, streamline data ingestion, and integrate with various cloud and database services, thereby accelerating the development of robust, data-driven AI applications.

List of Prompts

No prompt templates are mentioned in the provided repository or documentation.

List of Resources

No explicit resources are defined or exposed in the accessible repository content.

List of Tools

• list_sources: Lists available sources from the Unstructured API.
• get_source_info: Retrieves detailed information about a specific source connector.
• create_source_connector: Creates a new source connector.
• update_source_connector: Updates an existing source connector by parameters.
• delete_source_connector: Deletes a source connector by source ID.
• list_destinations: Lists available destinations from the Unstructured API.
• get_destination_info: Retrieves detailed information about a specific destination connector.
• create_destination_connector: Creates a destination connector by parameters.
• update_destination_connector: Updates an existing destination connector by ID.
• delete_destination_connector: Deletes a destination connector by ID.
• list_workflows: Lists workflows from the Unstructured API.
• get_workflow_info: Retrieves detailed information about a specific workflow.
• create_workflow: Creates a new workflow with source, destination ID, etc.
• run_workflow: Runs a specific workflow by workflow ID.
• update_workflow: Updates an existing workflow by parameters.
• delete_workflow: Deletes a specific workflow by ID.
• list_jobs: Lists jobs for a specific workflow.
• get_job_info: Retrieves detailed information about a specific job by job ID.
• cancel_job: Deletes (cancels) a specific job by ID.
• list_workflows_with_finished_jobs: Lists all workflows with completed jobs, including source and destination details.

Use Cases of this MCP Server

• Data Pipeline Automation: Streamline the setup and orchestration of complex ETL (Extract, Transform, Load) workflows by programmatically managing sources, destinations, and workflows.
• Connector Lifecycle Management: Automate the creation, updating, and deletion of connectors for popular cloud storage, database, and SaaS platforms (e.g., S3, Azure, Salesforce).
• Workflow Execution and Monitoring: Enable AI assistants to trigger, monitor, and manage jobs and workflows, ensuring smooth data operations and quick response to failures or status changes.
• Integration with Vector Databases: Seamlessly connect to vector databases such as Weaviate or Pinecone, enabling advanced AI applications requiring vector search capabilities.
• Data Governance and Auditing: Programmatically list, inspect, and audit all jobs and completed workflows to support compliance and data governance needs.

How to set it up

Windsurf

1. Ensure prerequisites like Python and relevant dependencies are installed.
2. Locate your Windsurf configuration file (e.g., windsurf.config.json).
3. Add the UNS-MCP server to the mcpServers section using the following JSON snippet:

   {
     "mcpServers": {
       "unstructured-mcp": {
         "command": "uns-mcp",
         "args": ["server"]
       }
     }
   }
   

4. Save the configuration and restart Windsurf.
5. Verify the UNS-MCP server appears as an available MCP server.

Claude

1. Locate your Claude desktop configuration file (e.g., claude_desktop_config.json).
2. Add the UNS-MCP server configuration as shown below:

   {
     "mcpServers": {
       "unstructured-mcp": {
         "command": "uns-mcp",
         "args": ["server"]
       }
     }
   }
   

3. Save the file and restart Claude.
4. Confirm setup by checking MCP server availability.

Cursor

1. Open your Cursor configuration (e.g., cursor.config.json).
2. Add the MCP server configuration:

   {
     "mcpServers": {
       "unstructured-mcp": {
         "command": "uns-mcp",
         "args": ["server"]
       }
     }
   }
   

3. Save changes and restart Cursor.
4. Validate the MCP server connection.

Cline

1. Open your Cline settings file.
2. Insert the following MCP server configuration:

   {
     "mcpServers": {
       "unstructured-mcp": {
         "command": "uns-mcp",
         "args": ["server"]
       }
     }
   }
   

3. Save and restart Cline.
4. Check for MCP server integration.

Securing API Keys

• Use environment variables to manage sensitive API keys and credentials.
• Example .env or environment specification:

  {
    "env": {
      "ANTHROPIC_API_KEY": "your-api-key",
      "AWS_KEY": "your-aws-key",
      "AWS_SECRET": "your-aws-secret",
      "WEAVIATE_CLOUD_API_KEY": "your-weaviate-api-key"
    },
    "inputs": {
      // Other tool-specific inputs
    }
  }
  

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "unstructured-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change "unstructured-mcp" to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Our opinion

The UNS-MCP server excels in tool coverage and setup documentation, but lacks explicit resource and prompt template exposure. It is highly practical for data pipeline management and connector automation, but could improve on MCP resource standardization and documentation.

MCP Score

Rating: 6/10 — The server is functional and well-documented for tool usage and connector management, but is missing key MCP features like prompt and resource definition, as well as licensing clarity. This reduces its utility for some advanced MCP workflows.