MCP Server Toolkit
Allows performing web searches using DuckDuckGo's API, returning results with titles, URLs, and snippets.
Enables querying SQLite databases with SELECT statements and listing tables.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP Server Toolkit show system info"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server Toolkit
A production-ready MCP server exposing filesystem, web search, SQLite, and system tools — plug directly into Claude Desktop or any MCP-compatible client.
What is MCP?
The Model Context Protocol is an open standard that lets AI assistants like Claude securely call external tools — giving them real-time access to your filesystem, databases, and the web without you having to paste content into the chat manually.
Related MCP server: Remote MCP Server
Features
Filesystem —
read_filereturns any text file with numbered lines;search_filesglobs a directory tree. Both tools enforce a configurableROOT_DIRso path-traversal attacks are impossible.Web Search —
web_searchqueries DuckDuckGo's free JSON API. No API key, no rate limits, async with a 10 s timeout.SQLite —
query_sqliteruns SELECT-only queries and returns typed columns + rows;list_tablesshows the schema at a glance.System —
get_system_infosnapshots OS, Python version, CPU count, memory usage, free disk, hostname, and uptime in one call.
Quick Start
Option 1 — pip
pip install git+https://github.com/plasmacat420/mcp-server-toolkit.git
mcp-toolkit # starts the server on stdio (Claude Desktop mode)Option 2 — Docker
docker pull ghcr.io/plasmacat420/mcp-server-toolkit:latest
docker run -it ghcr.io/plasmacat420/mcp-server-toolkit:latestOption 3 — Docker Compose (recommended for SSE / networked use)
git clone https://github.com/plasmacat420/mcp-server-toolkit
cd mcp-server-toolkit
cp .env.example .env # edit ROOT_DIR if needed
docker compose upThe SSE endpoint is then available at http://localhost:8000.
Claude Desktop Integration
Add the following block to claude_desktop_config.json
(~/Library/Application Support/Claude/ on macOS,
%APPDATA%\Claude\ on Windows):
{
"mcpServers": {
"mcp-toolkit": {
"command": "mcp-toolkit",
"env": {
"ROOT_DIR": "/Users/you/projects"
}
}
}
}Restart Claude Desktop — the four tool categories appear automatically in every conversation.
Tool Reference
Tool | Description | Parameters | Returns |
| Read a text file with line numbers |
|
|
| Glob-search inside a directory |
|
|
| DuckDuckGo search (no key needed) |
|
|
| Execute a SELECT query |
|
|
| List all tables in a SQLite DB |
|
|
| Host OS / resource snapshot | (none) |
|
CLI Usage
The mcp-client binary lets you call any tool from your terminal:
# Search for Python files
mcp-client search-files . "*.py"
# {"results": [...], "count": 12}
# Read a file
mcp-client read-file src/mcp_toolkit/server.py
# {"content": " 1 | \"\"\"MCP Server Toolkit...", "lines": 34, ...}
# Web search
mcp-client web-search "python asyncio tutorial" --max-results 3
# {"results": [{"title": "...", "url": "...", "snippet": "..."}], ...}
# Query a SQLite database
mcp-client query-db examples/sample.db "SELECT name, email FROM users LIMIT 3"
# {"columns": ["name", "email"], "rows": [["Alice Johnson", "alice@..."]], ...}
# System snapshot
mcp-client system-info
# {"os": "Linux", "cpu_count": 8, "memory_gb": 15.87, ...}
# Full demo against sample.db
mcp-client demoDevelopment
git clone https://github.com/plasmacat420/mcp-server-toolkit
cd mcp-server-toolkit
# Install with dev extras
pip install -e ".[dev]"
# Create the sample database
python examples/create_db.py
# Run the test suite
pytest -v
# Lint + format check
ruff check .
ruff format --check .
# Auto-fix
ruff check . --fix && ruff format .
# Full demo (requires sample.db)
python examples/demo.pyRunning a single test
pytest tests/test_filesystem.py::test_read_file_success -vArchitecture
The server is built on FastMCP, which handles the MCP wire protocol.
Each tool category lives in its own module under src/mcp_toolkit/tools/;
the modules export plain async functions that know nothing about MCP.
server.py creates the FastMCP instance, imports every tool function,
and registers them with mcp.tool(). Configuration is a single
pydantic-settings BaseSettings object (config.py) that reads from
environment variables or a .env file. The CLI client (client/cli.py)
imports the same async functions directly — no MCP protocol involved —
making it easy to smoke-test individual tools.
src/mcp_toolkit/
├── server.py ← FastMCP app + tool registration
├── config.py ← pydantic-settings Settings singleton
└── tools/
├── filesystem.py read_file, search_files
├── websearch.py web_search
├── database.py query_sqlite, list_tables
└── system.py get_system_infoLicense
MIT © plasmacat420
Available Tools
6 toolsget_system_infoA
Return a snapshot of the host system's resource usage and identity.
Returns:
Dict with keys os, os_version, python_version,
cpu_count, memory_gb, memory_used_pct,
disk_free_gb, hostname, and uptime_hours, or
{"error": message} on failure.
Inputs: no named parameters. Returns: object
list_tablesB
List all user-defined tables in a SQLite database.
Inputs: db_path (required). Returns: object
query_sqliteA
Execute a SELECT query against a SQLite database file.
Only SELECT statements are permitted; any other statement type returns an error without touching the database.
Inputs: db_path (required), sql (required). Returns: object
read_fileA
Read a text file and return its content with line numbers.
Inputs: path (required). Returns: object
search_filesB
Search for files matching a glob pattern inside a directory.
Inputs: directory (required), pattern (required), recursive. Returns: object
web_searchA
Search the web using DuckDuckGo's free API (no API key required).
Inputs: query (required), max_results. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool targets a clearly distinct domain: system info, SQLite tables, SQL queries, file reading, file search, and web search. No overlap in functionality.
All tools follow a consistent verb_noun pattern in snake_case (e.g., get_system_info, list_tables, query_sqlite, read_file, search_files, web_search), making predictions easy.
6 tools is a well-scoped number for a general-purpose toolkit, covering system, database, file, and web operations without being excessive or sparse.
The toolkit lacks write operations for files (e.g., write_file) and SQL modifications (insert/update/delete), limiting its usefulness for common tasks. This is a notable gap for a general toolkit.
Maintenance
Related MCP Connectors
Use your Mac, Windows or Linux computer from ChatGPT, Claude or Codex: files, commands, documents.
Work on your own computers from Claude or ChatGPT: run commands, edit and search files.
Safe folder access for ChatGPT and Claude: read, write and search files, risky tools opt-in.
Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables secure local filesystem operations and interactive terminal sessions for AI assistants. Provides 12 tools for file management, directory operations, code searching, and running interactive REPLs with security protections.34 npmMIT
- FlicenseNot gradedqualityDmaintenanceEnables AI tools like Claude to interact with a remote machine's file system and shell via a secure HTTPS endpoint. It provides standardized tools for executing shell commands, reading and writing files, and navigating directories.-
- AlicenseNot gradedqualityDmaintenanceProvides tools for AI-driven development workflows including file system operations, code analysis, code execution, web fetching, and search.Apache 2.0
- AlicenseAqualityDmaintenanceExposes local file I/O, a restricted shell whitelist, and SQLite helpers for safe integration with AI clients like Cursor and Claude Desktop.7MIT