mcpField
G
Guepard MCP SERVER

Guepard MCP SERVER

Local
github.com/Guepard-CorpDatabases0Updated 39w ago

About Guepard MCP SERVER

Guepard MCP Server exposes Guepard database-management APIs to teams managing deployments and related resources. Its broad tool set covers deployment operations, compute, users, logs, and metrics, with configurable tool selection.

Key features

  1. 01Manage deployments, branches, and snapshots
  2. 02Manage compute and performance profiles
  3. 03View logs and deployment metrics
  4. 04Manage database users, privileges, and tokens
  5. 05Select tools through predefined or custom configurations

Use cases

  1. 01Create and update Guepard deployments
  2. 02Create branches and snapshots for deployments
  3. 03Check compute status and deployment metrics
  4. 04Limit available tools to monitoring or read-only operations

MCP support

Exposes Guepard platform APIs as MCP tools, covering authentication, deployments, branches, snapshots, compute, users, logs, metrics, and other modules. Predefined configurations and environment variables can control which tools are active.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"guepard": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"access_token",
"nourk13/guepard-mcp-server:v2"
],
"env": {
"access_token": "<access_token>"
}
}
}
}
Run commanddocker run -i --rm -e access_token nourk13/guepard-mcp-server:v2
Environment variables to set
access_tokenSecret

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
ocinourk13/guepard-mcp-server:v2

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Guepard MCP Server - Complete API Implementation

A comprehensive Model Context Protocol (MCP) server for the Guepard database management platform, providing access to all Guepard APIs through organized modules.

🚀 Quick Start

Using Docker Hub Image
# Pull the image
docker pull mghassen/guepard-custom-mcp:latest

# Run the MCP server
docker run --rm -i -e ACCESS_TOKEN="your_guepard_token" mghassen/guepard-custom-mcp:latest
Using with Cursor

Add this to your Cursor MCP settings (~/.cursor/mcp.json):

{
  "mcpServers": {
    "guepard": {
      "command": "docker",
      "args": ["run", "--rm", "-i", "-e", "ACCESS_TOKEN=your_guepard_token", "mghassen/guepard-custom-mcp:latest"],
      "env": {}
    }
  }
}
Local Development
# Clone the repository
git clone <repository-url>
cd guepard-mcp-server

# Install dependencies
pip install -r requirements.txt

# Set environment variables
export ACCESS_TOKEN="your_guepard_token"
export GUEPARD_API_URL="https://api.guepard.run"
export GUEPARD_AUTH_API="https://auth.guepard.run"
export SUPABASE_ANON_KEY="your_supabase_key"

# Run the server
python main.py

🛠️ Available Tools

The MCP server provides comprehensive access to all Guepard APIs through organized modules:

🔐 Authentication
  • login_supabase - Login with Supabase authentication
  • refresh_token_supabase - Refresh access token with Supabase
  • logout_supabase - Logout and invalidate tokens with Supabase
🚀 Deployments
  • list_deployments - Get all deployments (with optional filtering)
  • create_deployment - Create a new database deployment
  • get_deployment - Get deployment details
  • update_deployment - Update deployment
  • delete_deployment - Delete deployment
🌿 Branches
  • list_branches - Get branches for deployment
  • create_branch_from_snapshot - Create a new branch from snapshot
  • update_branch - Update branch
🔄 Checkouts
  • checkout_branch - Checkout to a specific snapshot
  • checkout_snapshot - Checkout to a random snapshot from deployment
📸 Snapshots
  • create_snapshot - Create a snapshot
  • list_snapshots_deployment - Get snapshots for deployment
  • list_snapshots_branch - Get snapshots for branch
🏗️ Shadows
  • list_shadows_for_deployment - List shadow deployments for a specific deployment
  • list_all_shadows - List all shadow deployments
  • create_shadow - Create a new shadow deployment from a snapshot
🖥️ Nodes
  • list_nodes - Get all nodes
  • create_node - Create a new node
  • get_node - Get node details
⚡ Performance Profiles
  • list_performance_profiles - Get all performance profiles
  • create_performance_profile - Create performance profile
  • update_performance_profile - Update performance profile
  • apply_performance_profile - Apply performance profile to deployment
💻 Compute
  • get_compute_status - Get compute status
  • get_compute - Get compute information
  • start_compute - Start compute for a deployment
  • stop_compute - Stop compute for a deployment
📊 Logs & Metrics
  • get_deployment_logs - Get deployment logs
  • get_deployment_metrics - Get deployment metrics
👥 Database Users
  • list_database_users - Get database users for deployment
  • create_database_user - Create database user
  • update_database_user - Update database user
  • delete_database_user - Delete database user
  • grant_privileges - Grant privileges to database user
  • revoke_privileges - Revoke privileges from database user
  • list_user_privileges - List privileges for database user
🔑 Tokens
  • list_tokens - Get all tokens
  • generate_token - Generate new token
  • revoke_token - Revoke token
📡 Notifications & Subscriptions
  • subscribe_deployment - Subscribe to deployment notifications
  • unsubscribe_deployment - Unsubscribe from deployment notifications
  • list_subscriptions - List all deployment subscriptions
  • manage_subscriptions - Manage multiple deployment subscriptions
🛠️ Utilities
  • test_connection - Test connection to Guepard API
  • list_configurations - List available predefined configurations
  • get_configuration - Get current server configuration

📋 Prerequisites

  • Docker installed and running (for Docker usage)
  • Python 3.12+ (for local development)
  • Valid Guepard access token
  • Cursor IDE (for MCP integration)

🔧 Configuration

Environment Variables

Required:

  • ACCESS_TOKEN: Your Guepard platform access token

Optional:

  • GUEPARD_API_URL: Guepard API base URL (defaults to https://api.guepard.run)
  • GUEPARD_AUTH_API: Guepard auth API URL (defaults to https://auth.guepard.run)
  • SUPABASE_ANON_KEY: Supabase anonymous key for authentication
  • POSTGRES16_PROFILE_ID: Performance profile ID for PostgreSQL 16
  • POSTGRES17_PROFILE_ID: Performance profile ID for PostgreSQL 17

Tool Activation (New!):

  • GUEPARD_CONFIG: Use predefined configuration (minimal, read_only, production, development, monitoring)
  • GUEPARD_ENABLED_MODULES: Comma-separated list of modules to enable
  • GUEPARD_ENABLED_TOOLS: Comma-separated list of specific tools to enable
  • GUEPARD_DISABLED_TOOLS: Comma-separated list of tools to disable
  • GUEPARD_DISABLED_MODULES: Comma-separated list of modules to disable
Example Usage
# Test the connection (all tools enabled)
docker run --rm -i -e ACCESS_TOKEN="your_token" mghassen/guepard-custom-mcp:latest

# Use predefined configuration (minimal tools)
docker run --rm -i -e ACCESS_TOKEN="your_token" -e GUEPARD_CONFIG="minimal" mghassen/guepard-custom-mcp:latest

# Use predefined configuration (read-only)
docker run --rm -i -e ACCESS_TOKEN="your_token" -e GUEPARD_CONFIG="read_only" mghassen/guepard-custom-mcp:latest

# Use predefined configuration (production-safe)
docker run --rm -i -e ACCESS_TOKEN="your_token" -e GUEPARD_CONFIG="production" mghassen/guepard-custom-mcp:latest

# Custom configuration (specific tools only)
docker run --rm -i -e ACCESS_TOKEN="your_token" -e GUEPARD_ENABLED_TOOLS="test_connection,list_deployments,get_deployment,start_compute,stop_compute" mghassen/guepard-custom-mcp:latest

# In Cursor, ask:
# "Test my Guepard connection"
# "Get all my deployments"
# "Create a new deployment with PostgreSQL 17"
# "Start compute for deployment xyz"
# "Create a database user with specific privileges"
# "List available configurations"
# "Get current configuration"

⚙️ Tool Activation & Configuration

The Guepard MCP Server now supports selective activation of tools and modules, allowing you to build custom configurations for different use cases.

🚀 Predefined Configurations
ConfigurationDescriptionUse Case
minimalMinimal deployment management onlyBasic operations
read_onlyRead-only access to deployments and monitoringMonitoring dashboards
productionProduction environment without dangerous operationsProduction deployments
developmentFull development environment with all toolsDevelopment work
monitoringMonitoring and observability onlySystem monitoring
🛠️ Custom Configuration Examples
# Enable only specific modules
docker run --rm -i \
  -e ACCESS_TOKEN="your_token" \
  -e GUEPARD_ENABLED_MODULES="deployments,compute" \
  mghassen/guepard-custom-mcp:latest

# Enable only specific tools
docker run --rm -i \
  -e ACCESS_TOKEN="your_token" \
  -e GUEPARD_ENABLED_TOOLS="test_connection,list_deployments,get_deployment,start_compute,stop_compute" \
  mghassen/guepard-custom-mcp:latest

# Disable dangerous operations
docker run --rm -i \
  -e ACCESS_TOKEN="your_token" \
  -e GUEPARD_DISABLED_TOOLS="delete_deployment,delete_database_user,revoke_token" \
  mghassen/guepard-custom-mcp:latest
🔍 Configuration Verification

Use the built-in tools to verify your configuration:

  • list_configurations - List all available predefined configurations
  • get_configuration - Get current server configuration details

🏗️ Project Structure

guepard-mcp-server/
├── src/
│   └── guepard_mcp/
│       ├── auth/           # Authentication tools
│       ├── deployments/   # Deployment management
│       ├── branches/       # Branch management
│       ├── snapshots/      # Snapshot management
│       ├── shadows/        # Shadow deployment management
│       ├── nodes/          # Node management
│       ├── performance/    # Performance profiles
│       ├── compute/        # Compute management
│       ├── users/          # Database users
│       ├── tokens/         # Token management
│       ├── utils/          # Base classes and utilities
│       └── server.py       # Main server implementation
├── main.py                 # Entry point
├── requirements.txt        # Python dependencies
├── Dockerfile             # Docker configuration
└── README.md              # This file

🐳 Docker Images

  • mghassen/guepard-custom-mcp:latest - Latest version
  • mghassen/guepard-custom-mcp:2.0.0 - Stable version 2.0.0

🔍 Troubleshooting

Common Issues
  1. "ACCESS_TOKEN is required"

    • Make sure you've set the ACCESS_TOKEN environment variable
    • Verify your token is valid
  2. "No tools available in Cursor"

    • Restart Cursor after updating mcp.json
    • Check that Docker is running
    • Verify the image is pulled: docker pull mghassen/guepard-custom-mcp:latest
  3. Connection issues

    • Test the connection first: docker run --rm -i -e ACCESS_TOKEN="your_token" mghassen/guepard-custom-mcp:latest
    • Check your internet connection
    • Verify your Guepard credentials are valid
  4. Authentication errors

    • Ensure SUPABASE_ANON_KEY is set for auth operations
    • Verify GUEPARD_AUTH_API is correct
Debug Mode

Run with debug logging:

docker run --rm -i -e ACCESS_TOKEN="your_token" -e DEBUG=true mghassen/guepard-custom-mcp:latest

📚 API Reference

Tool Parameters
Authentication Tools
  • login_supabase: email (required), password (required)
  • refresh_token_supabase: refresh_token (required)
  • logout_supabase: refresh_token (required)
Deployment Tools
  • list_deployments: status (optional), limit (optional)
  • create_deployment: repository_name (required), performance_profile_id (required), plus optional fields
  • get_deployment: deployment_id (required)
  • update_deployment: deployment_id (required), plus fields to update
  • delete_deployment: deployment_id (required)
Branch Tools
  • list_branches: deployment_id (required)
  • create_branch_from_snapshot: deployment_id, branch_id, snapshot_id, branch_name (all required), is_ephemeral (optional)
  • update_branch: deployment_id, branch_id (required), plus fields to update
Checkout Tools
  • checkout_branch: deployment_id, branch_id, snapshot_id (all required)
  • checkout_snapshot: deployment_id (required), plus optional parameters
Snapshot Tools
  • create_snapshot: deployment_id, branch_id, snapshot_comment (all required)
  • list_snapshots_deployment: deployment_id (required)
  • list_snapshots_branch: deployment_id, branch_id (both required)
Shadow Tools
  • list_shadows_for_deployment: deployment_id (required)
  • list_all_shadows: No parameters
  • create_shadow: deployment_id, snapshot_id, repository_name, branch_name, performance_profile_id (all required)
Node Tools
  • list_nodes: No parameters
  • create_node: label_name, node_type, memory, cpu, storage (required), plus optional fields
  • get_node: node_id (required)
Performance Profile Tools
  • list_performance_profiles: No parameters
  • create_performance_profile: label_name, min_cpu, min_memory (required), plus optional fields
  • update_performance_profile: performance_profile_id (required), plus fields to update
  • apply_performance_profile: deployment_id, performance_profile_id (both required)
Compute Tools
  • get_compute_status: deployment_id (required)
  • get_compute: deployment_id (required)
  • start_compute: deployment_id (required), notify (optional)
  • stop_compute: deployment_id (required), notify (optional)
Logs & Metrics Tools
  • get_deployment_logs: deployment_id (required), lines (optional), level (optional), component (optional)
  • get_deployment_metrics: deployment_id (required), time_range (optional), metric_type (optional)
Database User Tools
  • list_database_users: deployment_id (required)
  • create_database_user: deployment_id, username, password (required), privileges (optional)
  • update_database_user: deployment_id, username (required), password (optional)
  • delete_database_user: deployment_id, username (required)
  • grant_privileges: deployment_id, username, privileges (all required)
  • revoke_privileges: deployment_id, username, privileges (all required)
  • list_user_privileges: deployment_id, username (both required)
Token Tools
  • list_tokens: No parameters
  • generate_token: name (required), expires_in (optional)
  • revoke_token: token_id (required)
Notification & Subscription Tools
  • subscribe_deployment: deployment_id (required)
  • unsubscribe_deployment: deployment_id (required)
  • list_subscriptions: No parameters
  • manage_subscriptions: action (required), deployment_ids (required)
Utility Tools
  • test_connection: No parameters
  • list_configurations: No parameters
  • get_configuration: No parameters

🆘 Support

  • Issues: Report issues on GitHub
  • Documentation: Check the main README.md
  • Guepard Platform: Visit guepard.run

📄 License

MIT License - see LICENSE file for details.


Ready to use! Pull the image and start managing your Guepard deployments with Cursor! 🎉

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.