STDF

Provides an MCP server to access and analyze STDF-V4 data from semiconductor testing for fast, streaming extraction and multi-site analysis.
  • python

1

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "henryxu21-stdf_mcp": {
      "command": "python",
      "args": [
        "-m",
        "stdf_mcp.simple_server"
      ]
    }
  }
}

You run an MCP server that gives large language models fast, streaming access to STDF-V4 data used in semiconductor testing. It enables metadata extraction, record retrieval, test analysis, and yield calculations, all in a scalable, multi-site capable environment.

How to use

You connect an MCP client to the STDF MCP Server to perform targeted data operations. Start by launching the local server, then issue requests through your client to extract metadata, fetch records, analyze tests, or compute yields. The server streams data efficiently, supports multi-site setups, and exports results as JSON, CSV, or Excel files for downstream workflows.

Common usage patterns include: extracting a quick file overview to validate a STDF-V4 file, pulling pointer (PTR) or functional (FTR) records for focused analysis, generating a per-unit yield CSV, and creating statistical summaries of test parameters. You can filter by test name, number, or suite, and apply regex when narrowing results. Export options help you integrate results into spreadsheets or data lakes.

How to install

# Clone the STDF MCP Server repository
git clone https://github.com/your-org/stdf-mcp.git
cd stdf-mcp

# Install the package in editable mode
pip install -e .

Additional setup and configuration

Start the MCP server using Python’s module entry point. This runs the Simple Server variant that listens for MCP protocol connections via standard input and output.

# Start the MCP server (stdin/stdout protocol)
python -m stdf_mcp.simple_server

Configuration for Claude Desktop

Add a server entry to Claude Desktop so your client can reach the STDF MCP Server. Use the following configuration, adjusting the path to your installation as needed.

{
  "mcpServers": {
    "stdf": {
      "command": "python",
      "args": ["-m", "stdf_mcp.simple_server"],
      "cwd": "/path/to/stdf-mcp"
    }
  }
}

Usage examples

Extract File Overview: instruct the server to read a STDF file and return lot, wafer, device, test dates, and file statistics.

Extract Yield Data: trigger per-unit yield extraction to generate a CSV with columns PART_ID, HW Bin, SW Bin.

Filter Parametric Data: request PTR records for a specific test to obtain filtered results and statistics.

Test Time Analysis: get per-site execution times and summary statistics exported as CSV.

Troubleshooting

If you see a missing module error for the MCP package, reinstall in editable mode to ensure your environment points at the local package.

Ensure the STDF file is valid STDF-V4; STDF-V3 is not supported and will fail validation.

Available tools

extract_file_overview

File summary, identifiers, and key stats returned as JSON for quick validation of a STDF-V4 file.

extract_test_configuration

Extract hardware setup and test environment details to understand how tests were run.

extract_yield_summary

Provide yield-related counts and yield percentage from test results.

extract_facility_details

Return manufacturing facility context and related details.

extract_program_metadata

Provide test program version and metadata information.

extract_test_parameters

List all test parameters with their limits and ranges.

extract_pir_prr_records

Retrieve Part Information/Results Records for a given site or file.

extract_ptr_records

Fetch Parametric Test Records with optional filtering by test name/number/suite and site.

extract_ftr_records

Retrieve Functional Test Records for analysis.

extract_hbr_records

Extract Hardware Bin Records to analyze binning results.

extract_sbr_records

Extract Software Bin Records for software-level results.

extract_sdr_records

Obtain Site Description Records to characterize sites.

extract_mrr_record

Access the Master Results Record summarizing the run.

extract_test_parameter_summary

Produce statistical summaries with test suite filtering and min/max/avg/stddev.

extract_filtered_data

Export filtered data to Excel with multi-site support.

extract_test_time

Analyze test execution times with site filtering and statistics.

extract_per_unit_yield

Export per-unit yield data to CSV with PART_ID, HW Bin, SW Bin columns.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational