Update to README.md with a reference to the official Reexpress AI MCP…

[allenschmaltz]

… Server

This adds a reference to the official Reexpress AI MCP Server in README.md.

Description

Server Details

The Reexpress MCP Server is a drop-in solution to add state-of-the-art statistical verification to your complex LLM pipelines, as well as your everyday use of LLMs for search and QA for software development and data science settings. It's the first reliable, statistically robust AI second opinion for your AI workflows.

In addition to providing you (the user) with a principled estimate of confidence in the output given your instructions, Claude (or other tool-calling LLMs) itself can use the verification output to progressively refine its answer, determine if it needs additional outside resources or tools, or has reached an impasse and needs to ask you for further clarification or information.

The current version works on Apple silicon Macs running macOS 15 (Sequoia). The server ensembles external LLM APIs with an on-device PyTorch model and an updatable local dense database, over which it calculates a well-calibrated Similarity-Distance-Magnitude estimator.

Motivation and Context

Given the wide-spread use of LLMs in real-world settings, a principled and robust approach for uncertainty quantification over the output of such models is critical. Here, the estimator is relative to the binary classification decision of whether the response addresses the user's question or instruction. This is useful, for example, to constrain hallucinations (up to a suitable probabilisitic threshold for a given task), and to route conditional branching decisions in complex agent-based pipelines.

How Has This Been Tested?

The MCP server has been thoroughly tested with Claude Sonnet 3.7 on Claude Desktop for macOS, as well as with VSCode Copilot. For typical settings, it is recommended to combine web search (or as applicable, a retrieval system) and extended thinking when calling the main Reexpress tool.

Breaking Changes

Adding the server requires a standard update to the MCP client configuration JSON (as in claude_desktop_config.json). We currently use conda, rather than uv, since the official release of the Faiss dependency is distributed through conda.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update

Checklist

• I have read the MCP Protocol Documentation
• My changes follows MCP security best practices
• I have updated the server's README accordingly
• I have tested this with an LLM client
• My code follows the repository's style guidelines
• New and existing tests pass locally
• I have added appropriate error handling
• I have documented all environment variables and configuration options

Additional context

The server has an Apache-2.0 license.

[allenschmaltz]

Let us know what additional information is needed. The Reexpress MCP server is a key component for building complex agentic/LLM pipelines.

[olaservo]

Thanks for your contribution to the servers list. This has been merged in this combined PR: #2075

This is a new process we're trying out, so if you see any issues feel free to re-open the PR and tag me.