Skip to content

03-tool-selection-scrape-map-crawl-search-extract.md

Latest commit

 

History

History
92 lines (66 loc) · 2.69 KB

03-tool-selection-scrape-map-crawl-search-extract.md

File metadata and controls

92 lines (66 loc) · 2.69 KB
layout default
title Chapter 3: Tool Selection: Scrape, Map, Crawl, Search, Extract
nav_order 3
parent Firecrawl MCP Server Tutorial

Chapter 3: Tool Selection: Scrape, Map, Crawl, Search, Extract

Welcome to Chapter 3: Tool Selection: Scrape, Map, Crawl, Search, Extract. In this part of Firecrawl MCP Server Tutorial: Web Scraping and Search Tools for MCP Clients, you will build an intuitive mental model first, then move into concrete implementation details and practical production tradeoffs.

Effective Firecrawl usage depends on selecting the right tool for each information-retrieval task.

Learning Goals

  • choose tools based on known vs unknown URL scope
  • combine tools for multi-step research tasks
  • avoid over-crawling when simpler methods suffice

Tool Selection Matrix

Task Type Preferred Tool
single known URL scrape
many known URLs batch scrape variants
discover URLs on a domain map
broad web discovery search
large site traversal crawl with strict limits
structured extraction extract with schema guidance

Source References

Summary

You now have a decision framework for tool selection that balances depth, cost, and speed.

Next: Chapter 4: Client Integrations: Cursor, Claude, Windsurf, VS Code

Source Code Walkthrough

src/types/fastmcp.d.ts

The ToolContext interface in src/types/fastmcp.d.ts handles a key part of this chapter's functionality:

      };

  export interface ToolContext<Session = unknown> {
    session?: Session;
    log: Logger;
  }

  export type ToolExecute<Session = unknown> = (
    args: unknown,
    context: ToolContext<Session>
  ) => unknown | Promise<unknown>;

  export class FastMCP<Session = unknown> {
    constructor(options: {
      name: string;
      version?: string;
      logger?: Logger;
      roots?: { enabled?: boolean };
      authenticate?: (
        request: { headers: IncomingHttpHeaders }
      ) => Promise<Session> | Session;
      health?: {
        enabled?: boolean;
        message?: string;
        path?: string;
        status?: number;
      };
    });

    addTool(tool: {
      name: string;
      description?: string;

This interface is important because it defines how Firecrawl MCP Server Tutorial: Web Scraping and Search Tools for MCP Clients implements the patterns covered in this chapter.

How These Components Connect

flowchart TD
    A[ToolContext]
Loading