Model Context Protocol (MCP) Servers

What is Model Context Protocol (MCP) Servers?

Model Context Protocol (MCP) Servers is a collection of 27 MCP servers developed by FeedMob. They provide structured data and context to AI models, enabling interaction with specific domains and data sources such as advertising platforms, internal operations, and content management.

How to use Model Context Protocol (MCP) Servers?

Install individual servers using the provided curl script, specifying the server key (e.g., sensor-tower-reporting). Each server has its own documentation and may require API credentials and environment configuration. Run curl -fsSL ... | bash -s -- --list to discover available server keys.

Key features of Model Context Protocol (MCP) Servers

• 27 ready-to-use MCP servers for diverse domains
• Advertising platform reporting integrations (17 servers)
• Internal operations and team tools
• Security and infrastructure capabilities (IP geolocation, fraud detection)
• Content and media management workflows
• Workflow automation via n8n nodes

Use cases of Model Context Protocol (MCP) Servers

• Retrieve advertising spend and performance data from multiple platforms
• Search, create, and update GitHub issues with team-based filtering
• Manage work journals and client contacts across Slack and HubSpot
• Perform batch IP geolocation and fraud detection checks
• Automate image cropping, watermarking, and uploads

FAQ from Model Context Protocol (MCP) Servers

How many MCP servers are included?

27 servers are available across advertising, operations, security, content, and workflow categories.

How do I install a specific server?

Use the curl install script with the server key, for example: curl -fsSL https://raw.githubusercontent.com/feed-mob/fm-mcp-servers/main/scripts/install.sh | bash -s -- sensor-tower-reporting.

Are there any prerequisites or dependencies?

Each server may require its own API credentials and environment configuration; refer to the individual server’s README for details.

Can I review the installer script before running it?

Yes, pipe the script to less: curl -fsSL https://raw.githubusercontent.com/feed-mob/fm-mcp-servers/main/scripts/install.sh | less.

What license is this project under?

This project is licensed under the MIT License.