How do I use Copilot MCP Server?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Copilot MCP Server explain Python comprehensions" 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.

Copilot MCP Server

by ohaiibuzzle

Overview Schema Source Related Servers Discussions Admin

Python

Local

Copilot MCP Server

This (horrible horrible) project provides MCP servers that wrap the copilot CLI so third-party MCP clients can invoke Copilot through tools.

Exposed tools

copilot_prompt: Runs copilot -p "<prompt>" in non-interactive mode.
- Supports session continuity:
  - session_id -> --session-id <id>
  - resume_session -> --resume=<id-or-name>
  - continue_session=true -> --continue
- Exactly one session mode can be selected per call.
copilot_command: Runs general copilot ... commands (interactive/server flags are blocked).

Tool return structured output with:

ok
command
exit_code (when executed)
stdout
stderr

Related MCP server: Copilot MCP Server

Requirements

copilot CLI installed and authenticated
Python 3.10+ (for Python server)

Optionally set COPILOT_BIN if copilot is not in PATH.

Python setup

git clone ...
uv init .
uv --directory . run copilot_mcp_server.py

Run:

copilot-mcp-server

Example MCP client config

Python server:

{
  "mcpServers": {
    "copilot-as-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "~/Copilot-as-MCP",
        "run",
        "copilot_mcp_server.py"
      ]
    }
  }
}

Session examples

Continue with a known session ID:
- copilot_prompt(prompt="Refine the previous answer", session_id="...")
Resume by session/task identifier:
- copilot_prompt(prompt="Continue this task", resume_session="...")
Continue most recent session:
- copilot_prompt(prompt="Continue", continue_session=true)

Available Tools

2 tools

copilot_commandB

Run a general Copilot command. Use args exactly like the shell command after copilot.

ParametersJSON Schema

Name	Required	Description	Default
`cwd`	No
`args`	Yes
`timeout_seconds`	No

Output Schema

ParametersJSON Schema

Name	Required	Description
No output parameters

TDQS

B3.3/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It only says it runs a command and how to format args, but does not mention potential side effects, output behavior, execution context, or safety implications of executing arbitrary Copilot commands.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences with no redundant information. It leads with the purpose ('Run a general Copilot command') and then provides the essential usage detail about args.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

This tool executes arbitrary Copilot commands, yet the description is minimal. It does not explain how output is returned, how errors or timeouts are handled, or how `cwd` affects execution. The output schema exists but the description lacks enough detail to fully understand the tool's behavior in context.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The description adds meaningful semantics for the `args` parameter by explaining its format, but it does not describe `cwd` or `timeout_seconds`. With 0% schema description coverage, the description only compensates for one of three parameters, leaving the other two to be inferred from their names and defaults.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states 'Run a general Copilot command' with a specific verb and resource, and the guidance to use args exactly like the shell command after `copilot` distinguishes this tool from the sibling `copilot_prompt` by framing it as a CLI command runner.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives a usage instruction for args format ('Use args exactly like the shell command after `copilot`'), but it does not explicitly state when to use this tool versus the alternative `copilot_prompt` or mention any exclusions. The usage context is implied rather than explicitly contrasted with alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

copilot_promptA

Run a non-interactive Copilot prompt (copilot -p ...) and return stdout/stderr/exit code.

ParametersJSON Schema

Name	Required	Default
`cwd`	No
`model`	No
`prompt`	Yes
`allow_all`	No
`extra_args`	No
`output_format`	No	text
`timeout_seconds`	No

Output Schema

ParametersJSON Schema

Name	Required	Description
No output parameters

TDQS

A3.5/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It discloses that execution is non-interactive and that stdout/stderr/exit code are captured and returned. However, it does not mention potential side effects, prerequisites like authentication, or that the prompt may trigger external service calls. This is a moderate level of transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, front-loaded sentence with no wasted words. It clearly states the action and the return value, making it easy to parse quickly.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Despite having an output schema, the description is too thin for a tool with 7 parameters and a sibling tool. It omits any information about when to use this over `copilot_command`, does not explain non-obvious parameters like `allow_all` or `extra_args`, and lacks any caveats about timeout or output format. The completeness is low.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate for unannotated parameters. It only implicitly covers the 'prompt' parameter by mentioning the `copilot -p` command; no explanation is given for cwd, model, allow_all, extra_args, output_format, or timeout_seconds. This is insufficient for a 7-parameter tool.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool runs a non-interactive Copilot prompt via `copilot -p` and returns stdout/stderr/exit code. It identifies the specific verb ('run'), resource ('Copilot prompt'), and distinguishes itself from the sibling tool by emphasizing 'non-interactive'.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage context by specifying 'non-interactive', which suggests using this tool when you want a one-shot prompt rather than an interactive session. However, it does not explicitly mention the sibling tool `copilot_command` or provide when-not-to-use guidance, so the usage guidance is only implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

2 tool updatesv0.1.0Aug 7, 2026
- First observedcopilot_command
- First observedcopilot_prompt

Deploy Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

B3.4/5.0

Scored 2026-08-07 06:05 across 2 tools

Disambiguation2/5

copilot_prompt and copilot_command overlap significantly since copilot_command can also run prompts, making it unclear which tool to use for a prompt task. The descriptions provide some clarity but do not fully resolve the boundary.

Naming Consistency5/5

Both tool names follow a consistent 'copilot_<mode>' pattern with snake_case, which is predictable and easy to remember. The naming is uniform and well-aligned with the server's purpose.

Tool Count3/5

With only 2 tools, the server is minimal but matches the narrow scope of wrapping the Copilot CLI. However, the presence of a general command makes the prompt-specific tool feel somewhat redundant, and the count is at the low end of reasonable.

Completeness3/5

The general copilot_command covers all CLI operations, avoiding dead ends, but the server lacks specialized tools for common Copilot workflows such as auth, session management, or structured prompt handling. The surface is functional yet sparse.

Maintenance

ActivitySlowing

ResponsivenessNo issues

Resources

GitHub Repository

Need Help?

Related Servers

Related MCP Connectors

Writingmate MCPOAuth
ai.writingmate
Use AI models for chat, image, and video generation from Claude Code and other MCP hosts.
AI & Machine Learning Image & Video Processing
1
llm-orchestration-agent-2
ai.getvda
MCP-Native LLM Orchestration Agent
Agent Orchestration AI & Machine Learning
1
mcp
com.skopik
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
Agent Orchestration Knowledge & Memory Search
2
ctx A
io.github.vanshulgoyal101
Turn a GitHub repo or docs site into agent-ready context: pack it or search it, over MCP.
RAG Systems Search Documentation Access
611

Related MCP Servers

GitHub Copilot CLI MCP Server
Developer Tools Command Line Code Execution
wminson
A
license
Not graded
quality
D
maintenance
Wraps the GitHub Copilot CLI to provide natural language command suggestions, explanations, and shell configuration for shell, git, and GitHub CLI commands through the Model Context Protocol.
Updated 9 months ago (2025-12-11 01:41 UTC)
1
MIT
Copilot MCP Server
Developer Tools Code Analysis Coding Agents
WillianPaiva
A
license
B
quality
D
maintenance
Integrates GitHub Copilot with MCP-compatible tools to provide AI-powered code assistance, including chat, code explanation, and reviews. It leverages existing GitHub CLI authentication to support multiple models like GPT-4o and Claude 3.5 Sonnet.
Updated a year ago (2025-07-04 21:04 UTC)
4
42
2
MIT
GitHub Copilot MCP Server
Developer Tools Code Analysis
Poorgramer-Zack
A
license
Not graded
quality
C
maintenance
Integrates GitHub Copilot CLI with MCP clients to offer various coding assistance tools including asking questions, explaining code, suggesting commands, debugging, refactoring, generating tests, and reviewing code.
Updated 5 months ago (2026-04-23 02:27 UTC)
14
8
MIT
copilot-mcp-server
Code Analysis Developer Tools Command Line
x51xxx
A
license
C
quality
D
maintenance
Connects MCP-enabled editors to GitHub Copilot CLI for non-interactive code analysis, batch processing, and code review.
Updated 9 months ago (2025-12-02 00:24 UTC)
10
19
5
MIT

Copilot MCP Server

Exposed tools

Requirements

Python setup

Example MCP client config

Session examples

Available Tools

Output Schema

TDQS

Output Schema

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

GitHub Copilot CLI MCP Server

Copilot MCP Server

GitHub Copilot MCP Server

copilot-mcp-server