SAGE

🧙 SAGE-MCP: Simple AI Guidance Engine for Claude - Universal AI assistant with smart mode selection, conversation continuity, and intelligent file handling

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 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": {
    "david-strejc-sage-mcp": {
      "command": "python",
      "args": [
        "/path/to/sage-mcp/server.py"
      ],
      "env": {
        "XAI_API_KEY": "xai-...",
        "DEFAULT_MODEL": "gpt-4o",
        "CUSTOM_API_KEY": "custom-key",
        "CUSTOM_API_URL": "http://localhost:11434",
        "GOOGLE_API_KEY": "AIza...",
        "OPENAI_API_KEY": "your-key",
        "DEFAULT_PROVIDER": "openai",
        "ANTHROPIC_API_KEY": "your-key",
        "DISALLOWED_MODELS": "gpt-4o-legacy",
        "OPENROUTER_API_KEY": "sk-or-...",
        "OPENAI_ALLOWED_MODELS": "gpt-4o,gpt-4o-mini"
      }
    }
  }
}

SAGE-MCP is a universal MCP server that turns Claude into a versatile development assistant. It automatically selects the right mode and model for coding, planning, debugging, and more, while preserving conversation context and handling files intelligently to keep your work organized and efficient.

How to use

Connect to the Sage MCP server from your MCP client to start interactive sessions. You can run multiple modes in the same conversation, switch modes as your task evolves, and rely on automatic context preservation between steps. Use the assistant to chat, analyze code, review design, debug issues, plan architecture, generate tests, refactor code, or think deeply about complex problems. You can also leverage smart file handling to embed full file content, generate summaries for large bases, or reference files by IDs. Real-time web search can be used to look up documentation, best practices, and current standards while you work.

How to install

Prerequisites: you need Python and Git installed on your machine. You will also configure API keys to enable access to AI providers.

Clone the repository, install dependencies, and set your API keys.

# Clone the repository
git clone https://github.com/david-strejc/sage-mcp
cd sage-mcp

# Install dependencies
pip install -r requirements.txt

Configure API keys in your environment. At minimum provide keys for the AI providers you plan to use and set a default model/provider as needed.

export OPENAI_API_KEY="your-key-here"
export ANTHOPIC_API_KEY="your-key-here"
export GOOGLE_API_KEY="your-key-here"
export OPENROUTER_API_KEY="your-key-here"

# Optional defaults
export DEFAULT_MODEL="gpt-4o"
export DEFAULT_PROVIDER="openai"

Additional configuration and usage notes

The server includes a preconfigured example for Claude Desktop integration. You can place the Sage MCP server configuration in your Claude Desktop settings to point at the local Python runtime and the Sage server entry point.

Environment variables control model access and behavior. You can restrict models, enable or disable web search, and tune thinking depth and temperatures by mode.

Available tools

sage

Universal SAGE tool that orchestrates mode selection, file handling, memory, and web search to deliver adaptive AI assistance across multiple modes

Built by

VeilStrat

AI signals for GTM teams