Gary

개인용 MCP 서버 프로젝트
  • python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "garyjeong-gary-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-v",
        "/Users/gary/Documents/workspace:/workspace:ro",
        "-v",
        "/Users/gary/.aws:/root/.aws:ro",
        "-v",
        "/Users/gary/.zshrc:/root/.zshrc:ro",
        "-v",
        "/Users/gary/Documents/workspace/gary-mcp/.env:/app/.env:ro",
        "-e",
        "WORKSPACE_PATH=/workspace",
        "-e",
        "AWS_PROFILE=jongmun",
        "-e",
        "SHELL_RC_PATH=/root/.zshrc",
        "gary-mcp-server"
      ],
      "env": {
        "AWS_PROFILE": "jongmun",
        "SHELL_RC_PATH": "/root/.zshrc",
        "WORKSPACE_PATH": "/workspace"
      }
    }
  }
}

You run a flexible MCP server that unifies multiple specialized services into a single integration point. It lets you reference project documents, manage AWS and Fly.io resources, convert Markdown to PDF, and analyze code flows, all through a centralized, extensible MCP platform.

How to use

To use Gary MCP Server, connect your MCP client and select Gary MCP as your active MCP source. You can run multiple independent MCP services in parallel and consolidate their tools under one unified interface. You will access project documents, AWS/Fly.io resources, and code analysis capabilities through the integrated server, enabling automated document referencing, infrastructure queries, and code understanding within your workspace.

How to install

Prerequisites you need before installing Gary MCP Server are Python 3.12 or newer, Docker, and the AWS CLI configured with your jongmun profile. Optionally install the Fly.io CLI if you plan to manage Fly.io resources.

Install using Docker Compose (recommended)

# Start all MCP servers
docker-compose -f docker-compose.mcp.yml up -d

# Start only specific servers
docker-compose -f docker-compose.mcp.yml up -d aws-mcp db-mcp

# Stop all servers
docker-compose -f docker-compose.mcp.yml down

Install using individual Docker images

You can build and run each MCP server independently. Example commands for the AWS server show how to mount your host configuration and provide necessary environment settings.

# AWS server build and run
docker build -f docker/aws-cli-mcp.Dockerfile -t aws-mcp .
docker run -it --rm \
  -v ~/.aws:/root/.aws:ro \
  -v ~/.zshrc:/root/.zshrc:ro \
  -e AWS_PROFILE=jongmun \
  -e SHELL_RC_PATH=/root/.zshrc \
  aws-mcp
  • Volumes and environment: The host workspace is mounted read-only at /workspace, AWS settings at /root/.aws, and your shell rc at /root/.zshrc to auto-load AWS/Fly.io tokens. Use WORKSPACE_PATH to reference the workspace inside the container and SHELL_RC_PATH for the CLI shell configuration.
## Local development without Docker

uv installed

pip install uv

install dependencies

uv pip install -e .

run the server

python -m src.server

## Environment and configuration notes

Configure the following to ensure the server can source your workspace, AWS, and Fly.io credentials. You may mount host files and set environment variables as shown in the examples above.

## Cursor IDE 연동

To integrate Gary MCP Server with Cursor IDE, add one of the MCP server configurations to your Cursor MCP settings. Use the Docker-based approach for stability or run the server locally in Python/uv as shown.

## Available tools

### read\_document

Reads a document file from the workspace to enable contextual analysis and reference.

### list\_workspace\_projects

Scans the workspace for project directories to provide an overview.

### search\_documents

Searches the workspace documents for a keyword, optionally within a specific project.

### aws\_cli\_execute

Executes AWS CLI commands using the jongmun profile to query resources.

### aws\_list\_resources

Lists AWS resources for a given service and type.

### aws\_get\_account\_info

Retrieves information about the AWS account.

### flyio\_list\_apps

Lists apps deployed on Fly.io.

### flyio\_get\_app\_status

Gets the status of a Fly.io app.

### flyio\_get\_app\_logs

Fetches logs for a Fly.io app.

### github\_cli\_execute

Executes GitHub CLI commands directly.

### markdown\_to\_pdf

Converts Markdown documents to PDF with optional CSS styling.

### analyze\_code\_flow

Analyzes code flow within a project to identify structure and entry points.

### find\_related\_code

Finds related code segments based on functions, classes, or imports.

### get\_code\_reusability

Evaluates code reuse opportunities across a project.

### list\_databases

Lists databases and connection options available to the MCP.

### describe\_tables

Describes database table schemas.

### run\_query

Executes SQL queries with optional read-only or read-write modes.

### sync\_official\_docs

Synchronizes official documentation locally for offline use.

### list\_official\_docs

Lists documents currently mirrored locally.

### search\_official\_docs

Searches mirrored official documentation.

### resolve\_library\_id

Resolves library identifiers to IDs and metadata.

### list\_libraries

Lists supported libraries by category.

### get\_library\_docs

Fetches library documentation in Context7 style.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational