FireScrape

Provides autonomous research, web access, and local context tools for AI agents via an MCP server.

typescript

1

GitHub Stars

typescript

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": {
    "aasm3535-firescrape-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@yutugyutugyutug/firescrape-mcp"
      ]
    }
  }
}

FireScrape MCP Server provides a dedicated MCP server to empower AI agents with autonomous research, web access, and local project context. It bridges AI models with real-world tools, enabling precise web scraping, multi-step research, and seamless local file access for coding tasks.

How to use

You connect to FireScrape MCP from your MCP client to access a set of tools that let your AI agents conduct deep research, search the web, scrape pages into clean Markdown, discover other MCP servers, and read files from your local project. Use the stdio-based connection by starting the MCP server locally or connect via an HTTP-based client if supported. The server exposes a consistent tool surface that your agent can invoke to perform tasks, verify facts, and synthesize results. To start using it, launch the server with the provided command and integrate it into your agent’s MCP workflow. When your agent needs to fetch current information, scrape a page, or read a local file, invoke the corresponding tool through the MCP interface and process the structured output.

How to install

npx -y @yutugyutugyutug/firescrape-mcp

Additional sections

Prerequisites: you need Node.js and npm to use the stdio-based start command via npx. Ensure you have a working Node environment before attempting to launch the MCP server.

Available tools

deep_research

Conducts autonomous, multi-step research on complex topics, filtering noise and synthesizing reports.

search_web

Performs high-availability web searches with fallback providers to ensure results.

scrape_webpage

Converts messy HTML into clean, token-efficient Markdown, removing ads and irrelevant boilerplate.

discover_tools

A Meta-Tool to find and install other MCP servers from GitHub and Smithery.

list_files

Lists files in a folder to provide local project context.

read_file

Reads the content of a file to supply local information to the agent.

get_agent_rules

Returns the strict operational protocol for AI agents.