Snippets

MCP server for storing, searching, and managing code snippets using AI-powered semantic search and traditional keyword matching. Just ask your coding agent to save or search a certain snippet. That's it.

javascript

GitHub Stars

javascript

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": {
    "freakynit-snippets-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@freakynit/snippets-mcp@latest"
      ],
      "env": {
        "SNIPPETS_FILE_PATH": "Optional path to save snippets and embeddings in.. should have .json extension"
      }
    }
  }
}

Snippets MCP is a server for storing, searching, and managing code snippets using semantic search and traditional keyword matching. It helps you save, organize, and retrieve code snippets with fast, meaningful results, so you can quickly reuse ideas across projects.

How to use

You interact with Snippets MCP through your MCP client to add, search, and manage snippets. Start by saving snippets you frequently reuse, then search by intent or exact terms. Use semantic search to find snippets by meaning, and combine it with keyword filtering to narrow results by language or tags.

Typical usage patterns you can follow: 1) Add a snippet with its code, optional language, tags, and a short description to improve semantic search. 2) Search for related code by describing your goal in natural language or by keywords. 3) Filter results by language and tags to quickly find the most relevant snippet. 4) Update a snippet if its content or description changes, which also refreshes its embeddings for accurate future searches. 5) Delete a snippet you no longer need.

How to install

Prerequisites: you need a working Node.js environment. Install Node.js from the official source and verify Node.js and npm are available on your system.

Install the Snippets MCP package in your project or workspace.

Configure and run Snippets MCP through your MCP setup using the provided commands and environment variable. The following MCP configurations show how to run Snippets MCP as a local (stdio) server via your MCP runner.

Configuration and running notes

The Snippets MCP server can be run via MCP by adding a stdio configuration that executes the package with npx and passes the correct arguments. You can also run it via a Windows command shell if you are on Windows. The environment variable SNIPPETS_FILE_PATH controls where the JSON database with snippets and embeddings is stored.

Environment variable to configure: SNIPPETS_FILE_PATH

Purpose: path to save snippets and embeddings in a JSON file
Default: ~/.snippets-mcp-db.json if not provided

Examples and gotchas

If you want to run Snippets MCP via MCP, use the following configurations in your MCP setup. The HTTP and STDIO examples below show how to wire Snippets MCP into your environment.

Available tools

add-snippet

Adds a new code snippet to the database with code content, optional tags, language, and description to improve semantic search.

search-snippets

Searches for snippets using a hybrid of semantic similarity and keyword matching, with optional filters for tags, language, and date range.

update-snippet

Updates an existing snippet and re-generates embeddings if the code, tags, or description change.

delete-snippet

Deletes a snippet by its ID from the database.

get-snippet

Retrieves a single snippet by its ID.

Built by

VeilStrat

AI signals for GTM teams