How do I use mcp-upstage?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-upstage extract text from this PDF file" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

mcp-upstage

Official

by UpstageAI

Overview Schema Related Servers Score Discussions

Python

Remote

Upstage MCP

A Model Context Protocol (MCP) server for Upstage AI's document digitization and information extraction capabilities

Overview

The Upstage MCP Server provides a robust bridge between AI assistants and Upstage AI’s powerful document processing APIs. This server enables AI models—such as Claude—to effortlessly extract and structure content from various document types including PDFs, images, and Office files. The package supports multiple formats and comes with seamless integration options for Claude Desktop.

Related MCP server: MCP-Upstage-Server

Key Features

Document Digitization: Extract structured content from documents while preserving layout.
Information Extraction: Retrieve specific data points using intelligent, customizable schemas.
Multi-format Support: Handles JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, and XLSX.
Claude Desktop Integration: Effortlessly connect with Claude and other MCP clients.

Prerequisites

Before using this server, ensure you have the following:

Upstage API Key: Obtain your API key from Upstage API.
Python 3.10+: The server requires Python version 3.10 or higher.
The MCP server relies upon Astral UV to run, please install

Installation & Configuration

This guide provides step-by-step instructions to set up and configure the mcp-upstage

Using `uv` (Recommended)

No additional installation is required when using uvx as it handles execution. However, if you prefer to install the package directly:

uv pip install mcp-upstage

Configure Claude Desktop

For integration with Claude Desktop, add the following content to your claude_desktop_config.json:

Configuration Location

Windows: %APPDATA%\Claude\claude_desktop_config.json
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Using uvx Command (Recommended)

{
  "mcpServers": {
    "mcp-upstage": {
      "command": "uvx",
      "args": ["mcp-upstage"],
      "env": {
        "UPSTAGE_API_KEY": "<your-api-key>"
      }
    }
  }
}

If uvx is not available globally, you may encounter a Server disconnected error. To resolve this, run which uvx to find its full path, and replace "command": "uvx" above with the returned path.

After adding the configuration, restart Claude Desktop to apply the changes.

Output Directories

Processing results are stored in your home directory under:

Document Parsing Results:
~/.mcp-upstage/outputs/document_parsing/
Information Extraction Results:
~/.mcp-upstage/outputs/information_extraction/
Generated Schemas:
~/.mcp-upstage/outputs/information_extraction/schemas/

Local/Development Setup

Follow these steps to set up and run the project locally:

Step 1: Clone the Repository

git clone https://github.com/UpstageAI/mcp-upstage.git
cd mcp-upstage

Step 2: Set Up the Python Environment

# Install uv if not already installed
pip install uv

# Create and activate a virtual environment
uv venv

# Activate the virtual environment
# On Windows:
# .venv\Scripts\activate
# On macOS/Linux:
source .venv/bin/activate

# Install dependencies in editable mode
uv pip install -e .

Step 3: Configure Claude Desktop for Local Testing

Download Claude Desktop:
Download Claude Desktop

Open and Edit Configuration:

Navigate to Claude → Settings → Developer → Edit Config.
Edit the claude_desktop_config.json file with the following configurations:

For Windows:

{
  "mcpServers": {
    "mcp-upstage": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "C:\\path\\to\\cloned\\mcp-upstage",
        "python",
        "-m",
        "upstage_mcp.server"
      ],
      "env": {
        "UPSTAGE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Replace C:\\path\\to\\cloned\\mcp-upstage with your actual repository path.

For macOS/Linux:

{
  "mcpServers": {
    "mcp-upstage": {
      "command": "/Users/username/.local/bin/uv",
      "args": [
        "run",
        "--directory",
        "/path/to/cloned/mcp-upstage",
        "python",
        "-m",
        "upstage_mcp.server"
      ],
      "env": {
        "UPSTAGE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Replace:

/Users/username/.local/bin/uv with the output of which uv.
/path/to/cloned/mcp-upstage with the absolute path to your local clone.

Tip for macOS/Linux users: If connection issues occur, using the full path to your uv executable can improve reliability.

After configuring, restart Claude Desktop.

Available Tools

The server exposes two primary tools for AI models:

Document Parsing (parse_document):
- Description: Processes documents and extracts content while preserving structure.
- Parameter:
  file_path – the path to the document to be processed.
- Example Query:
  "Can you parse the document at C:\Users\username\Documents\contract.pdf and provide a summary?"
Information Extraction (extract_information):
- Description: Extracts structured information from documents based on predefined or auto-generated schemas.
- Parameters:
  file_path – the document file path;
  schema_path (optional) – a JSON file with an extraction schema;
  auto_generate_schema (default true) – whether to auto-generate a schema.
- Example Query:
  "Extract the invoice number, date, and total from C:\Users\username\Documents\invoice.pdf."

Below is the revised troubleshooting section formatted as requested. You can copy and paste the following Markdown directly into your README:

Troubleshooting

Common Issues

API Key Missing:
Ensure that your UPSTAGE_API_KEY is correctly set in your claude_desktop_config.json file. Obtain a valid API key from Upstage Console.
File Not Found:
Double-check the file path for correctness and accessibility. Ensure that file paths are absolute (e.g., C:\Users\name\Documents\file.pdf) and that any special characters in the path are properly escaped.
Server Not Starting:
Verify that your virtual environment is activated and all dependencies are installed. Additionally, review the Claude Desktop log files for errors:
- Windows: %APPDATA%\Claude\logs\mcp-upstage.log
- macOS: ~/Library/Logs/Claude/mcp-upstage.log
Server Connection Issues:
Restart Claude Desktop. Ensure that uvx is installed and available in your system PATH, or use its absolute path in your configuration if needed.
Processing Failures:
Check that the document is in a supported format (PDF, JPEG, PNG, TIFF, etc.), its file size is under 50MB, and it contains fewer than 100 pages. Test with a simpler document to confirm functionality.
Invalid Document Format:
Verify that the document is in a supported, uncorrupted format.
Failed to Connect to Upstage API:
Confirm your network connection, firewall settings, and configuration details in claude_desktop_config.json. Review the logs for more detailed error messages.

Log Files

For troubleshooting, view the server logs at:

Windows:
%APPDATA%\Claude\logs\mcp-upstage.log
macOS:
~/Library/Logs/Claude/mcp-upstage.log

Contributing

Contributions are welcome! If you wish to enhance the project or add new features, please fork the repository and submit a pull request. For major changes, please open an issue first to discuss what you would like to change.

License

This project is licensed under the MIT License.

Available Tools

2 tools

extract_informationC

Extract structured information from documents using Upstage Universal Information Extraction.

This tool can extract key information from any document type without pre-training.
You can either provide a schema defining what information to extract, or let the system
automatically generate an appropriate schema based on the document content.

Supported file formats: JPEG, PNG, BMP, PDF, TIFF, HEIC, DOCX, PPTX, XLSX
Max file size: 50MB
Max pages: 100

Args:
    file_path: Path to the document file to process
    schema_path: Optional path to a JSON file containing the extraction schema
    schema_json: Optional JSON string containing the extraction schema
    auto_generate_schema: Whether to automatically generate a schema if none is provided

ParametersJSON Schema

Name	Required	Description
`file_path`	Yes	Path to the document file to process
`schema_json`	No	JSON string containing the extraction schema (optional)
`schema_path`	No	Path to JSON file containing the extraction schema (optional)
`auto_generate_schema`	No	Whether to automatically generate a schema

TDQS

C2.9/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It discloses supported file formats, maximum file size (50MB), and maximum pages (100), which are useful constraints. However, it does not mention whether the operation is read-only, what happens on failure, or any rate limits. Given that the tool is non-destructive, the transparency is adequate but not comprehensive.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is reasonably structured with a brief intro followed by bullet points for constraints and a list of arguments. However, it is somewhat verbose, especially the 'Args' section which duplicates the input schema. Not all sentences earn their place; the 'without pre-training' phrase is extraneous.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has 4 parameters, no output schema, and no annotations, the description should cover output format and error behavior. It does not explain what the extracted information looks like or how to interpret the result. It also does not mention any prerequisites (e.g., valid document content) or edge cases, leaving significant gaps for the agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description adds context that the schema can be provided via file path, JSON string, or auto-generated, which clarifies the relationship between the parameters. However, it largely repeats the schema descriptions without adding significant new meaning beyond explaining the optionality.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool extracts structured information from documents, using a specific technology (Upstage Universal Information Extraction). It mentions key capabilities like schema provision and auto-generation. However, it does not explicitly differentiate itself from the sibling tool 'parse_document', which could cause confusion about which tool to use for what.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explains the flexibility of providing a schema or auto-generating one, but it lacks guidance on when to use this tool versus its sibling 'parse_document'. There are no 'when to use' or 'when not to use' statements, nor prerequisites or context about the expected input document types.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

parse_documentB

Parse a document using Upstage AI's document digitization API.

This tool extracts the structure and content from various document types,
including PDFs, images, and Office files. It preserves the original formatting
and layout while converting the document into a structured format.

Supported file formats include: PDF, JPEG, PNG, TIFF, and other common document formats.

ParametersJSON Schema

Name	Required	Description	Default
`file_path`	Yes	Path to the document file to be processed
`output_formats`	No	Output formats (e.g., 'html', 'text', 'markdown')

TDQS

B3.1/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden of behavioral disclosure. It only mentions high-level behavior (extracting, preserving formatting) and supported formats, but fails to disclose authentication needs, rate limits, side effects (e.g., if files are modified), or the exact nature of the structured output. This is insufficient for a tool of this complexity.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is relatively concise with three short paragraphs. The first sentence is a clear purpose statement. The rest adds useful detail about format preservation and supported types. It is well-structured and not overly verbose, though minor repetition exists.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity and lack of output schema, the description is incomplete. It does not mention capabilities like OCR, table extraction, language support, or output structure details. While it covers basic functionality, an agent would need more context for proper invocation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3. The description adds value by listing supported file formats beyond what the schema provides, and by explaining that formatting and layout are preserved, which gives context to the output. It does not, however, elaborate on the 'output_formats' parameter beyond the schema's examples.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool parses documents using Upstage AI API, extracting structure and content while preserving formatting. It mentions supported file formats, giving a good sense of the tool's purpose. However, it does not explicitly distinguish from the sibling tool 'extract_information', so it loses a point for full clarity.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives. It does not mention prerequisites, ideal use cases, or when to avoid it. The sibling tool 'extract_information' is listed but not contrasted, leaving the agent to infer usage context without help.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

2 tool updatesv0.1.0Jul 30, 2026
- First observedextract_information
- First observedparse_document

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

B3.4/5.0

Disambiguation5/5

The two tools have clearly distinct purposes: parse_document extracts layout-preserving content, while extract_information extracts structured data using a schema. No overlap in functionality.

Naming Consistency5/5

Both tool names use the verb_noun pattern with snake_case (parse_document, extract_information), maintaining consistent naming conventions.

Tool Count3/5

With only 2 tools, the server feels minimal for a document processing domain. While focused, it is borderline thin for typical workflows.

Completeness4/5

The two tools cover core document digitization and extraction needs. Minor gaps exist (e.g., no file upload or conversion), but the primary use cases are addressed.

Maintenance

ActivityInactive

ResponsivenessSyncing

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Brainiall Documents A
com.brainiall
High-fidelity PDF to structured Markdown conversion and document field extraction.
Data Platforms
31
Iteration Layer AOAuth
com.iterationlayer
Composable APIs for document extraction, image transformation, and document & sheet generation.
App Automation Image & Video Processing Multimedia Processing
2
extend
ai.extend
Turn documents into structured data: parse, extract, classify, split, and fill PDF forms.
1
Rosetta PDF
io.github.Jcjimenezglez
Parse, extract, split, and ask over digital PDFs (text layer, no OCR) from Cursor and Claude.
File Systems Documentation Access AI & Machine Learning
1

Related MCP Servers

Claude RAG MCP Pipeline
RAG Systems Vector Databases Search
kenjisekino
A
license
Not graded
quality
D
maintenance
Enables Claude Desktop to search and query personal document collections (PDF, Word, Markdown, text) using semantic search and conversational AI with full context preservation across exchanges.
Updated a year ago (2025-09-20 18:06 UTC)
MIT
MCP-Upstage-Serverofficial
Documentation Access Research & Data Data Platforms
UpstageAI
A
license
A
quality
D
maintenance
Enables document processing through Upstage AI services including parsing various document formats, extracting structured information with custom schemas, auto-generating extraction schemas, and classifying documents into categories.
Updated 7 months ago (2026-02-11 17:36 UTC)
4
29
3
MIT
Upstage MCP Server
AI & Machine Learning Data Platforms
PritamPatil2603
A
license
A
quality
C
maintenance
Enables AI assistants to extract and structure content from documents (PDFs, images, Office files) using Upstage AI's document digitization and information extraction APIs.
Updated a year ago (2025-04-17 03:52 UTC)
2
3
MIT
Enterprise Knowledge MCP
RAG Systems Search
j84077200345-dotcom
A
license
Not graded
quality
B
maintenance
Enables querying enterprise documents (DOCX, PDF, PPTX) using natural language, with hybrid search and MCP integration for Claude Desktop and other agents.
Updated 3 months ago (2026-06-20 14:57 UTC)
MIT

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/UpstageAI/mcp-upstage'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

Upstage MCP

Overview

Key Features

Prerequisites

Installation & Configuration

Using uv (Recommended)

Configure Claude Desktop

Configuration Location

Using uvx Command (Recommended)

After adding the configuration, restart Claude Desktop to apply the changes.

Output Directories

Local/Development Setup

Step 1: Clone the Repository

Step 2: Set Up the Python Environment

Step 3: Configure Claude Desktop for Local Testing

Available Tools

Troubleshooting

Common Issues

Log Files

Contributing

License

Available Tools

TDQS

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

Claude RAG MCP Pipeline

MCP-Upstage-Serverofficial

Upstage MCP Server

Enterprise Knowledge MCP

Latest Blog Posts

MCP directory API

Using `uv` (Recommended)