duckcoop
MCP server for sandboxed, read-only SQL queries on CSV/Parquet/JSON files via DuckDB, limited to a specified directory.
README
duckcoop
MCP server that lets an LLM client run sandboxed, read-only SQL (via DuckDB) against CSV/Parquet/JSON files under a directory you choose. Like a duck, it only gets to paddle around the pen you build for it.
Usage
npm run build
node dist/cli.js /path/to/sandbox-root
Optional flags: --scratch-dir <path> (defaults to a temp directory), --max-temp-size <size> (defaults to 4GB).
Exposes two tools: query (SELECT / CREATE TEMP VIEW / CREATE TEMP TABLE only) and list_files (data files under the sandbox root). See docs/adr/ for the security model.
Registering with an MCP client
Claude Code
claude mcp add duckcoop node /absolute/path/to/duckcoop/dist/cli.js /absolute/path/to/sandbox-root
Claude Desktop
Add an entry to claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"duckcoop": {
"command": "node",
"args": [
"/absolute/path/to/duckcoop/dist/cli.js",
"/absolute/path/to/sandbox-root"
]
}
}
}
Restart Claude Desktop for the change to take effect.
Querying files you drag-and-drop into a chat: Claude Desktop doesn't save chat attachments to a path duckcoop can see. Files dropped in a Local Agent Mode session instead land under ~/Library/Application Support/Claude/local-agent-mode-sessions/<account-id>/<session-id>/local_<id>/outputs/. Point the sandbox root at the stable parent directory, ~/Library/Application Support/Claude/local-agent-mode-sessions, so newly created session folders are covered without editing the config again:
{
"mcpServers": {
"duckcoop": {
"command": "node",
"args": [
"/absolute/path/to/duckcoop/dist/cli.js",
"/Users/<you>/Library/Application Support/Claude/local-agent-mode-sessions"
]
}
}
}
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.