unichatLABSCO SUMMARY
A second model behind one tool — swap providers by changing two environment variables.
The point is provider-independence: the same unichat tool fronts OpenAI, Anthropic or any compatible endpoint, so getting a different model's take is a config change rather than a different server. The four prompts are the practical add-on — code review, documentation, explanation and rework as ready-made instructions rather than tools.
What it is
A server that exposes a single tool for chatting with a configured model, plus a set of code-focused prompts, across many providers.
What you get
- unichat — one tool that sends messages to the configured model and returns a response
- A code_review prompt that reviews code for issues and improvements
- A document_code prompt that generates docstrings and comments
- An explain_code prompt that walks through how a piece of code works
- A code_rework prompt that applies requested changes to code
- One configuration reaches OpenAI, Anthropic, and OpenAI-compatible providers such as MistralAI, xAI, Google AI, DeepSeek and Alibaba
Requirements
A vendor API key. Set UNICHAT_MODEL to the model and UNICHAT_API_KEY to its key; for an OpenAI-compatible provider, add UNICHAT_BASE_URL pointing at its endpoint, and the built-in model list is skipped. Run it with uvx and the unichat-mcp-server package.
Setup effort
Add the configuration to your client
USE CASES
01
Getting a second model's opinion mid-conversation
Put the question to the configured model and read its answer in the same thread.
02
Reviewing a piece of code
Run the code_review prompt to get issues and improvements flagged.
03
Documenting code you just wrote
Use the document_code prompt to generate docstrings and comments.
04
Pointing the same setup at a cheaper provider
Change UNICHAT_MODEL, UNICHAT_API_KEY and UNICHAT_BASE_URL to move to an OpenAI-compatible endpoint.
1 tool
Listed in project docs. Your client may show a different list.
BEFORE YOU INSTALL
What to know first
Tools you get
1 tool named in the project's documentation.
Requires
- uvx on PATH
- UNICHAT_API_KEY in your environment
INSTALLATION
Connect your client
MCP Clients
Claude Desktop — uvx
CLAUDE DESKTOP — UVX · JSON
{
"mcpServers": {
"unichat-mcp-server": {
"command": "uvx",
"args": [
"unichat-mcp-server"
],
"env": {
"UNICHAT_MODEL": "<YOUR_UNICHAT_MODEL>",
"UNICHAT_API_KEY": "<YOUR_UNICHAT_API_KEY>"
}
}
}
}Claude Desktop — local build
Get the source code
CLAUDE DESKTOP — LOCAL BUILD · TERMINAL
git clone https://github.com/amidabuddha/unichat-mcp-server.gitAdd this to claude_desktop_config.json. <YOUR_SOURCE_CODE_LOCAL_DIRECTORY> is the full path of the folder you ran git clone in, the one that holds unichat-mcp-server.
CLAUDE DESKTOP — LOCAL BUILD · JSON
{
"mcpServers": {
"unichat-mcp-server": {
"command": "uv",
"args": [
"--directory",
"<YOUR_SOURCE_CODE_LOCAL_DIRECTORY>/unichat-mcp-server",
"run",
"unichat-mcp-server"
],
"env": {
"UNICHAT_MODEL": "<YOUR_UNICHAT_MODEL>",
"UNICHAT_API_KEY": "<YOUR_UNICHAT_API_KEY>"
}
}
}
}
