Unified AI Access Point via MCP in Python

This repository hosts the Python implementation for interfacing with numerous prominent commercial and open-source large language models through a unified Model Context Protocol (MCP) abstraction layer.

Also accessible in a TypeScript variant: TypeScript Implementation

This service relays inputs to OpenAI, MistralAI, Anthropic, xAI, Google AI, DeepSeek, Alibaba, and Inception backends, leveraging the standardized MCP communication mechanism, either through direct tool invocation or predefined contextual templates. Note: Configuration requires corresponding vendor API credentials.

Available Functionality (Tooling)

The gateway exposes a primary interface tool: - unichat: Executes a generalized query against the selected backend model. - Parameter: messages (String, mandatory input containing conversation history or query). - Output: The model's generated response.

Integrated Workflow Templates (Prompts)

Pre-packaged operations leveraging the underlying models: - code_review - Purpose: Assess submitted code for adherence to established best practices, identify potential defects, and suggest architectural improvements. - Required Argument: - code (String): The source code block slated for analysis. - document_code - Purpose: Automatic generation of comprehensive explanatory documentation, including inline comments and formal docstrings. - Required Argument: - code (String): The source code requiring annotation. - explain_code - Purpose: Provide a thorough, step-by-step breakdown of the functionality embodied within a segment of programming logic. - Required Argument: - code (String): The code snippet necessitating explanation. - code_rework - Purpose: Modify supplied source code based on specified transformation instructions. - Arguments: - changes (String, optional): Detailed description of modifications to be implemented. - code (String, mandatory): The original source code base to be transformed.

Initialization Guide

Setup for Local Execution

Claude Desktop Integration

Configuration file locations: On macOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

Model Selection: Refer to this external resource for the complete list of valid model identifiers available for the "SELECTED_UNICHAT_MODEL" setting. Ensure that "YOUR_UNICHAT_API_KEY" is correctly mapped to the appropriate vendor credentials.

Configuration Example for Environment Variables:

"env": { "UNICHAT_MODEL": "gpt-4o-mini", "UNICHAT_API_KEY": "YOUR_OPENAI_API_KEY" }

Development/Local Server Launch Configuration:

"mcpServers": { "unified-ai-mcp-gateway": { "command": "uv", "args": [ "--directory", "{{your source code local directory}}/unichat-mcp-server", "run", "unichat-mcp-server" ], "env": { "UNICHAT_MODEL": "SELECTED_UNICHAT_MODEL", "UNICHAT_API_KEY": "YOUR_UNICHAT_API_KEY" } } }

Published/System Deployment Configuration:

"mcpServers": { "unified-ai-mcp-gateway": { "command": "uvx", "args": [ "unichat-mcp-server" ], "env": { "UNICHAT_MODEL": "SELECTED_UNICHAT_MODEL", "UNICHAT_API_KEY": "YOUR_UNICHAT_API_KEY" } } }

Automated Installation via Smithery

To deploy this gateway automatically for use with the Claude Desktop client using Smithery:

bash npx -y @smithery/cli install unichat-mcp-server --client claude

Development Lifecycle

Building Artifacts and Distribution

Procedure for preparing the software package for release:

1. Clean up previous build outputs: bash rm -rf dist

2. Synchronize project dependencies and update the resolution file: bash uv sync

3. Execute the build process to generate distributable archives: bash uv build

Results will reside in the dist/ folder (source and wheel distributions).

1. Upload to the Python Package Index (PyPI): bash uv publish --token {{YOUR_PYPI_API_TOKEN}}

Troubleshooting Execution

Debugging processes that communicate over standard input/output streams (stdio) can be complex. For the superior debugging experience, leverage the official MCP Inspector.

Launch the Inspector via npm using the following command structure:

bash npx @modelcontextprotocol/inspector uv --directory {{your source code local directory}}/unichat-mcp-server run unichat-mcp-server

Once initiated, the Inspector will present a network address. Access this address in a web browser to commence interactive debugging sessions.

WIKIPEDIA: Cloud computing is defined by ISO as "a paradigm for enabling network access to a scalable and elastic pool of shareable physical or virtual resources with self-service provisioning and administration on-demand." It is often abbreviated as "the cloud".

== Core Attributes == In 2011, the US National Institute of Standards and Technology (NIST) established five critical hallmarks of cloud architectures. These are the precise stipulations from NIST:

On-demand self-service: "A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service provider." Broad network access: "Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations)." Resource pooling: " The provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand." Rapid elasticity: "Capabilities can be elastically provisioned and released, in some cases automatically, to scale rapidly outward and inward commensurate with demand. To the consumer, the capabilities available for provisioning often appear unlimited and can be appropriated in any quantity at any time." Measured service: "Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service. By 2023, the International Organization for Standardization (ISO) had expanded and refined the list.

== Historical Context ==

The conceptual lineage of cloud computing traces back to the 1960s, when the notion of time-sharing gained traction through Remote Job Entry (RJE). The dominant operational structure during that period was the 'data center' approach, where users submitted batch jobs to dedicated operators who executed them on mainframe systems. This era was characterized by intense investigation into methodologies for democratizing access to high-capacity computation via time-sharing, focusing on optimizing the underlying infrastructure, platform layers, and application layers, thereby boosting end-user efficiency. The 'cloud' visual shorthand for abstracted, virtualized services originated in 1994. It was employed by General Magic to describe the conceptual 'domain' accessible to mobile agents operating within their Telescript environment. The attribution for this metaphor generally goes to David Hoffman, a General Magic communications professional, who adapted it from its established usage in telecommunications and networking contexts. The phrase 'cloud computing' gained significantly wider recognition in 1996, appearing in a business strategy document drafted by Compaq Computer Corporation concerning the future trajectory of computation and the Internet. The company's initial objective was to superch