Skip to main content الرئيسية المنشئون jeremylongshore tons-of-skills-marketplace firecrawl-data-handling
firecrawl-data-handling Process, validate, and store Firecrawl scraped content with deduplication and chunking.
Use when handling scraped markdown, implementing content pipelines, building RAG knowledge
bases, or processing crawl results for downstream consumption.
Trigger with phrases like "firecrawl data", "firecrawl content processing",
"firecrawl markdown cleaning", "firecrawl storage", "firecrawl RAG pipeline".
الانتقال إلى التثبيت سوق المهارات اكتشف واستكشف مهارات الذكاء الاصطناعي التي بناها المجتمع.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
عرض تفاصيل Prompt يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/jeremylongshore/tons-of-skills-marketplace --skill firecrawl-data-handlingيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
تحميل Zip جاري التحميل... المزيد من هذا المستودع langchain-deploy-integration Deploy a LangChain 1.0 / LangGraph 1.0 app to Cloud Run, Vercel, or LangServe correctly — with timeouts sized for chain length, cold-start mitigation, SSE anti-buffering headers, and Secret Manager over .env. Use when prepping a first production deploy, debugging a stream that hangs behind a proxy, or diagnosing p99 latency spikes. Trigger with "langchain deploy", "langchain cloud run", "langchain vercel python", "langchain langserve", or "langchain docker".
langchain-langgraph-agents Build a correct LangGraph 1.0 ReAct agent with create_react_agent — typed tools, error propagation, recursion caps, and stop conditions that actually stop. Use when writing a first tool-calling agent, migrating from AgentExecutor or initialize_agent, or diagnosing an agent that loops on vague prompts. Trigger with "langgraph agent", "create_react_agent", "langgraph tool calling", "AgentExecutor migration", or "agent loop cost".
langchain-langgraph-human-in-loop Build LangGraph 1.0 human-in-the-loop approval flows with interrupt_before /
interrupt_after and Command(resume=...) — JSON-serializable state, clean
resume semantics, and UI wiring for approval decisions. Use when adding an
approval gate before an expensive tool call, wiring a Slack/web UI for agent
approvals, or debugging a graph that crashes on interrupt.
Trigger with "langgraph human in loop", "langgraph interrupt_before",
"langgraph approval flow", "Command resume", "langgraph HITL".
name firecrawl-data-handling description Process, validate, and store Firecrawl scraped content with deduplication and chunking.
Use when handling scraped markdown, implementing content pipelines, building RAG knowledge
bases, or processing crawl results for downstream consumption.
Trigger with phrases like "firecrawl data", "firecrawl content processing",
"firecrawl markdown cleaning", "firecrawl storage", "firecrawl RAG pipeline".
allowed-tools Read, Write, Edit version 1.11.0 license MIT author Jeremy Longshore <jeremy@intentsolutions.io> tags ["saas","firecrawl","compliance"] compatibility Designed for Claude Code
Firecrawl Data Handling
Overview
Process scraped web content from Firecrawl pipelines. Covers markdown cleaning, structured data extraction with Zod validation, content deduplication, chunking for LLM/RAG, and storage patterns for crawled content.
Instructions
Step 1: Content Cleaning
import FirecrawlApp from "@mendable/firecrawl-js" ;
const firecrawl = new FirecrawlApp ({
apiKey : process.env .FIRECRAWL_API_KEY !,
});
async function scrapeClean (url : string const result = await firecrawl.scrapeUrl (url, {
formats : ["markdown" ],
onlyMainContent : true ,
excludeTags : ["script" , "style" , "nav" , "footer" , "iframe" ],
waitFor : 2000 ,
});
return {
url : result.metadata ?.sourceURL || url,
title : result.metadata ?.title || "" ,
markdown : cleanMarkdown (result.markdown || "" ),
scrapedAt : new
Date
toISOString
function
cleanMarkdown
md : string
string
return
replace
/\n{3,}/g
"\n\n"
replace
/\[.*?\]\(javascript:.*?\)/g
""
replace
/!\[.*?\]\(data:.*?\)/g
""
replace
/<!--[\s\S]*?-->/g
""
replace
/<script[\s\S]*?<\/script>/gi
""
trim
Step 2: Structured Extraction with Validation import { z } from "zod" ;
const ArticleSchema = z.object ({
title : z.string ().min (1 ),
author : z.string ().optional (),
publishedDate : z.string ().optional (),
content : z.string ().min (50 ),
wordCount : z.number (),
});
async function extractArticle (url : string const result = await firecrawl.scrapeUrl (url, {
formats : ["extract" ],
extract : {
schema : {
type : "object" ,
properties : {
title : { type : "string" },
author : { type : "string" },
publishedDate : { type : "string" },
content : { type : "string" },
},
required : ["title" , "content" ],
},
},
});
if (!result.extract ) throw new Error (`Extraction failed for ${url} ` );
return ArticleSchema .parse ({
...result.extract ,
wordCount : (result.extract .content || "" ).split (/\s+/ ).length ,
});
}
Step 3: Content Deduplication import { createHash } from "crypto" ;
function contentHash (text : string string {
return createHash ("sha256" )
.update (text.trim ().toLowerCase ())
.digest ("hex" );
}
function deduplicatePages (pages : Array <{ url: string ; markdown: string }>const seen = new Map <string , string >();
const unique : typeof pages = [];
const duplicates : Array <{ url : string ; duplicateOf : string }> = [];
for (const page of pages) {
const hash = contentHash (page.markdown );
if (seen.has (hash)) {
duplicates.push ({ url : page.url , duplicateOf : seen.get (hash)! });
} else {
seen.set (hash, page.url );
unique.push (page);
}
}
console .log (`Dedup: ${pages.length} input, ${unique.length} unique, ${duplicates.length} duplicates` );
return { unique, duplicates };
}
Step 4: Chunk for LLM / RAG interface ContentChunk {
url : string ;
title : string ;
chunkIndex : number ;
content : string ;
wordCount : number ;
}
function chunkForRAG (
url : string ,
title : string ,
markdown : string ,
maxWords = 800
): ContentChunk [] {
const sections = markdown.split (/\n(?=#{1,3}\s)/ );
const chunks : ContentChunk [] = [];
let current = "" ;
let index = 0 ;
for (const section of sections) {
const combined = current ? `${current} \n\n${section} ` : section;
if (combined.split (/\s+/ ).length > maxWords && current) {
chunks.push ({
url, title, chunkIndex : index++,
content : current.trim (),
wordCount : current.split (/\s+/ ).length ,
});
current = section;
} else {
current = combined;
}
}
if (current.trim ()) {
chunks.push ({
url, title, chunkIndex : index,
content : current.trim (),
wordCount : current.split (/\s+/ ).length ,
});
}
return chunks;
}
Step 5: Crawl and Store Pipeline import { writeFileSync, mkdirSync } from "fs" ;
import { join } from "path" ;
async function crawlAndStore (baseUrl : string , outputDir : string , opts ?: {
maxPages?: number ;
paths?: string [];
}mkdirSync (outputDir, { recursive : true });
const crawlResult = await firecrawl.crawlUrl (baseUrl, {
limit : opts?.maxPages || 50 ,
includePaths : opts?.paths ,
scrapeOptions : { formats : ["markdown" ], onlyMainContent : true },
});
const pages = (crawlResult.data || []).map (page =>url : page.metadata ?.sourceURL || baseUrl,
markdown : cleanMarkdown (page.markdown || "" ),
}));
const { unique } = deduplicatePages (pages);
const manifest = unique.map (page =>const slug = new URL (page.url ).pathname
.replace (/\//g , "_" ).replace (/^_|_$/g , "" ) || "index" ;
const filename = `${slug} .md` ;
writeFileSync (join (outputDir, filename), page.markdown );
return { url : page.url , file : filename, size : page.markdown .length };
});
writeFileSync (join (outputDir, "manifest.json" ), JSON .stringify (manifest, null , 2 ));
return manifest;
}
Error Handling Issue Cause Solution Empty content JS not rendered Increase waitFor, use onlyMainContent Garbage in markdown Bad HTML cleanup Add excludeTags for problematic elements Duplicate pages URL aliases or redirects Content-hash deduplication Oversized chunks Long single sections Add word limit to chunking logic Extract returns null Page too complex for LLM Simplify schema, use shorter prompt
Examples
Documentation Scraper with RAG Output const docs = await crawlAndStore ("https://docs.example.com" , "./scraped-docs" , {
maxPages : 50 ,
paths : ["/docs/*" , "/api/*" ],
});
for (const doc of docs) {
const content = readFileSync (`./scraped-docs/${doc.file} ` , "utf-8" );
const chunks = chunkForRAG (doc.url , doc.file , content);
console .log (`${doc.url} : ${chunks.length} chunks` );
}
Resources
Next Steps For access control, see firecrawl-enterprise-rbac.
