SMRT Docs

Routes SMRT package documentation to synthesize AI-backed responses with package references.

typescript

GitHub Stars

typescript

Language

6 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": {
    "happyvertical-smrt-docs-mcp": {
      "command": "pnpm",
      "args": [
        "exec",
        "tsx",
        "/path/to/smrt/packages/smrt-mcp/src/index.ts"
      ],
      "env": {
        "DEBUG": "false",
        "HAVE_AI_TYPE": "openai",
        "GEMINI_API_KEY": "your Gemini key",
        "OPENAI_API_KEY": "sk-...",
        "HAVE_AI_API_KEY": "your-api-key",
        "ANTHROPIC_API_KEY": "your-anthropic-key"
      }
    }
  }
}

You run a Documentation MCP Server that routes your questions to the SMRT package CLAUDE.md knowledge bases. It orchestrates package-specific documentation, synthesizes an answer with AI, and references the sources it used. This enables you to answer complex developer questions by leveraging built-in package expertise.

How to use

You interact with the MCP server through a client that sends your questions to the SMRT advisor. The server selects relevant packages based on the keywords in your query, loads their CLAUDE.md documentation, and uses AI to synthesize a final answer that includes package references.

In practice, you use the advisor to get concise, source-backed guidance for SMRT packages. When your question touches multiple packages, expect the answer to cite those packages explicitly and point to the underlying CLAUDE.md content for deeper reading.

How to install

Prerequisites: ensure you have Node.js and a package manager installed on your system.

Install the MCP server package using your preferred package manager.

Run the MCP server as described in the configuration snippet shown below.

pnpm install @happyvertical/smrt-mcp

Configuration and running

Configure your MCP client to start the server and expose it as an MCP endpoint. Use the stdio-based local server configuration that runs the MCP server process.

The following example shows how to wire the server as an MCP entry named smrt_advisor. It runs via pnpm and executes the TypeScript entry point for the MCP server.

{
  "mcpServers": {
    "smrt_advisor": {
      "type": "stdio",
      "command": "pnpm",
      "args": [
        "exec",
        "tsx",
        "/path/to/smrt/packages/smrt-mcp/src/index.ts"
      ],
      "env": {
        "DEBUG": "false"
      }
    }
  }
}

AI provider and environment variables

To power the ask tool, configure an AI provider in your environment. You may supply any of the supported providers by setting one or more of these keys.

Possible environment keys include OpenAI, Anthropic, and Gemini keys. If you do not provide an AI provider, you can still use other tools like listing packages or retrieving docs.

Security and best practices

Keep your AI keys and DEBUG settings secure. Do not commit sensitive keys to version control. Restrict access to the MCP endpoint to trusted clients.

Troubleshooting

If you encounter AI initialization or connectivity issues, verify you have a valid AI provider key configured. If you cannot locate relevant packages, use the list-packages tool to explore available options and then request information for the specific packages.

Available tools

ask

Ask a question about SMRT and automatically routes to relevant packages and synthesizes a response.

list-packages

List all available SMRT framework packages with descriptions and keywords.

get-docs

Fetch the full CLAUDE.md documentation for a specific package.

Built by

VeilStrat

AI signals for GTM teams