TextIn xParse MCP Server
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@TextIn xParse MCP Serverparse document /documents/report.pdf"
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.
TextIn xParse MCP Server
SSE MCP server for TextIn xParse document parsing.
The server exposes the four Python SDK methods documented by TextIn:
parse_run: maps toclient.parse.run()parse_create_job: maps toclient.parse.create_job()parse_get_job: maps toclient.parse.get_job()parse_wait_job: maps toclient.parse.wait_job()
Setup
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
pip install -e .Configure authentication with environment variables only:
$env:TEXTIN_APP_ID="your-app-id"
$env:TEXTIN_SECRET_CODE="your-secret-code"Do not pass credentials as MCP tool arguments. The server constructs XParseClient() without explicit credentials so the official SDK reads its environment variables.
To use a custom TextIn-compatible server address, set TEXTIN_SERVER_URL:
$env:TEXTIN_SERVER_URL="https://your-server.example.com"When TEXTIN_SERVER_URL is unset, the official SDK default server is used.
The MCP tools accept file content as base64. Decoded file size is limited by MAX_FILE_BYTES, which defaults to 52428800 bytes.
Related MCP server: Mineru MCP Server
Run
textin-mcpThe default transport is SSE. You can also pass an explicit transport:
textin-mcp --transport sseThe server listens on 127.0.0.1:8000 by default. Override this for container or remote access:
$env:MCP_HOST="0.0.0.0"
$env:MCP_PORT="8000"
textin-mcp --transport sseDocker
Build the image:
docker build -t textin-mcp .Run the SSE MCP server:
docker run --rm -p 8000:8000 `
-e TEXTIN_APP_ID="your-app-id" `
-e TEXTIN_SECRET_CODE="your-secret-code" `
-e TEXTIN_SERVER_URL="https://your-server.example.com" `
-e MAX_FILE_BYTES="52428800" `
textin-mcpOr run with Docker Compose:
docker compose up --buildCompose reads TEXTIN_APP_ID, TEXTIN_SECRET_CODE, optional TEXTIN_SERVER_URL, and optional MAX_FILE_BYTES from your environment or .env file. It starts the MCP SSE service on host port 8000 and the file-to-base64 helper on host port 8005.
File to Base64 Helper
The helper is a regular HTTP service, not an MCP SSE endpoint. It accepts a multipart file upload and returns the payload needed by parse_run or parse_create_job.
Run locally:
textin-file-base64By default it listens on 0.0.0.0:8001. Override with:
$env:FILE_BASE64_HOST="0.0.0.0"
$env:FILE_BASE64_PORT="8001"
textin-file-base64Request:
curl.exe -X POST "http://127.0.0.1:8005/file-to-base64" `
-F "file=@D:\docs\example.pdf"Response:
{
"filename": "example.pdf",
"mime_type": "application/pdf",
"size": 12345,
"base64": "JVBERi0x..."
}Tools
parse_run
Synchronously parse a local document.
Parameters:
filename: document filename, for exampleexample.pdffile_base64: base64-encoded document contentpage_range: optional page range, for example1-10password: optional encrypted PDF passwordinclude_hierarchyinclude_inline_objectsinclude_char_detailsinclude_image_datainclude_table_structurepagestitle_treetable_view:markdownorhtml
parse_create_job
Create an asynchronous parsing job.
Parameters:
filename: document filename, for exampleexample.pdffile_base64: base64-encoded document contentwebhook: optional completion callback URLall parse configuration parameters supported by
parse_run
parse_get_job
Query an asynchronous parsing job.
Parameters:
job_id
parse_wait_job
Poll until an asynchronous parsing job finishes.
Parameters:
job_idtimeout: default300.0poll_interval: default5.0download_result: when true, downloadsresult_urlafter completion
MCP Client Example
{
"mcpServers": {
"textin-xparse": {
"url": "http://127.0.0.1:8000/sse"
}
}
}Reference
TextIn Python SDK documentation: https://docs.textin.com/xparse/v1/sdk-python
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
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/aenron/textin-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server