mcp-github-copilot
Allows querying any GitHub Copilot model programmatically, with support for file attachments and model discovery.
Click on "Deploy 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., "@mcp-github-copilot Explain async/await using Copilot"
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.
MCP server for GitHub Copilot — query any Copilot model programmatically via the official SDK.
Website · Documentation
Features
No extra API key — uses your existing GitHub Copilot CLI credentials automatically.
Any model — target GPT-5, Codex, Claude Sonnet, or any model your subscription grants.
File and image attachments — attach local files or base64 blobs alongside a prompt.
Model discovery — list available models with context window limits and billing multipliers.
Streaming progress — sends MCP progress notifications for each streamed chunk.
Related MCP server: BMAD MCP Server
Install
npm install -g @kud/mcp-github-copilotUsage
Add the server to your MCP client configuration:
{
"mcpServers": {
"mcp-github-copilot": {
"command": "mcp-github-copilot"
}
}
}Tools
Tool | Description |
| Send a prompt to a Copilot model and return the response. Accepts an optional |
| List all available Copilot models with capabilities, context window limits, and billing multipliers. |
Development
git clone https://github.com/kud/mcp-github-copilot.git
cd mcp-github-copilot
npm install
npm run dev📚 Full documentation → mcp-github-copilot/docs
Available Tools
2 toolslist_modelsA
List available Copilot models.
Inputs: no named parameters.
queryA
Send a prompt to GitHub Copilot and return the response. Uses logged-in Copilot CLI credentials automatically.
Inputs: prompt (required), model, attachments.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools are completely distinct: one sends prompts to Copilot, the other lists available models. There is no overlap or ambiguity between them.
The naming is inconsistent: 'query' is a single-word verb while 'list_models' follows a verb_noun pattern. The generic 'query' also lacks specificity compared to 'list_models'.
With only two tools, the server feels minimal but borderline appropriate for a basic Copilot integration. It covers essential actions but leaves little room for a richer workflow.
The server covers the core action of querying Copilot and listing models, but lacks features like streaming, conversation management, or model details, which may be expected in a fuller implementation.
Related MCP Connectors
Shared memory MCP server for Claude Code, Codex, Gemini CLI and Cursor. Cited, editable.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
MCP server for OpenAI API (chat completions, image generation, embeddings) via AceDataCloud
Driflyte MCP server which lets AI assistants query topic-specific knowledge from web and GitHub.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceA modular MCP server that extends GitHub Copilot's capabilities through intelligent context compression and dynamic model routing for long-lived coding sessions.4-
- FlicenseNot gradedqualityDmaintenanceA Model Context Protocol (MCP) server that bridges BMAD agents with GitHub Copilot, enabling AI-assisted development workflows through specialized agent tools and prompts.1-
- AlicenseAqualityDmaintenanceWraps GitHub Copilot CLI as an MCP server, enabling MCP clients to run Copilot conversations for code tasks with support for session resumption and permission modes.4MIT
- AlicenseAqualityBmaintenanceMCP server for opencode that queries GitHub Copilot models via a persistent opencode server, requiring no API keys.288 npm3MIT