gbrigandi/mcp-server-cortex

Overview

A Rust-based server that implements the Model Context Protocol (MCP) for use with Cortex, a local AI inference engine. It allows AI tools and assistants to interact with Cortex models through a standardized protocol. The server is open source and designed for lightweight, local deployment.

Best for

Best for
Developers building local AI toolchains with Cortex who need a lightweight MCP adapter

Use cases

• Connect AI assistants to local Cortex models via MCP
• Build custom tool integrations for Cortex-based workflows
• Enable local AI inference in development environments

How to use

Tools exposed

• analyze_ip_with_abuseipdb
• analyze_with_abusefinder
• scan_url_with_virustotal
• analyze_url_with_urlscan_io

Tested with

Claude Desktop

Example client config

{\n  "mcpServers": {\n    "cortex": {\n      "command": "/opt/mcp-servers/mcp-server-cortex",\n      "args": [],\n      "env": {\n        "CORTEX_ENDPOINT": "http://your-cortex-instance:9000/api",\n        "CORTEX_API_KEY": "your_cortex_api_key_here"\n      }\n    }\n  }\n}

Notes

A Rust-based server that implements the Model Context Protocol (MCP) for use with Cortex, a local AI inference engine. It allows AI tools and assistants to interact with Cortex models through a standardized protocol. The server is open source and designed for lightweight, local deployment.

15 stars on GitHub. Last updated 2025-12-06. Licensed MIT.

Use cases

• Connect AI assistants to local Cortex models via MCP
• Build custom tool integrations for Cortex-based workflows
• Enable local AI inference in development environments

Pros

• Written in Rust for performance and low resource usage
• Open source with a permissive license
• Simple protocol implementation for easy integration

Cons

• Very early stage with only 15 GitHub stars
• Limited community support and documentation
• Requires familiarity with both MCP and Cortex

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.