500 free credits. No credit card required.

One API for your AI agents’ web workflows

Give your AI agents scraping (without getting blocked), search, extraction, and verification in one toolkit. Use MCP for one-command setup, or use the API directly with your own key.

No credit card required
Up and running withOpenClawin 2 minutes
🦞
Clawie
OpenClaw Agent
Message...
Works with
OpenClaw
Claude Code
ChatGPT
Claude Desktop
Cursor
01

A Toolkit Your Agents Compose Freely

webscraperapi gives AI agents search, scraping, extraction, and verification as composable tools. They figure out how to chain them based on what you ask for.

View all tools →
🦞
Clawie
OpenClaw Agent

What tech stack does linear.app use?

2:34 PM

🔧webscraperapi_detect_technologies
url: "https://linear.app"
Next.js, React, Vercel, Cloudflare, Segment
🔧webscraperapi_scrape
url: "https://linear.app" output: "markdown"
3,847 chars extracted

Linear uses Next.js with React on Vercel, behind Cloudflare CDN. They also use Segment for analytics. The site is well-optimized — clean markup, fast load times.

2:35 PM

What tech stack does linear.app use?

Your agent, where you already are

Ask a question in Telegram, and your OpenClaw agent uses webscraperapi to research, scrape, and analyze — then replies with the answer. No dashboards, no context switching.

Agent builds a lead list

Your agent searches Google, scrapes company pages, and verifies founder emails — all through composable API calls. No orchestration code needed.

Claude
Find 10 fintech companies using React and get founder emails
I'll search Google, scrape each result, then verify emails.
Tool Call
GET /v2/google_search?query=fintech+companies+using+React
10 results: ["stripe.com", "plaid.com", "brex.com", ...]
Tool Call
GET /v2/scrape?url=stripe.com&output=markdown
# Stripe - Financial infrastructure for the internet...
Tool Call
GET /v2/verify_email_address?email_address=founder@stripe.com
{ "deliverable": true }
Found 10 companies with verified founder emails. Here's your list...
ChatGPT — competitor_monitor
Compare my store prices vs competitor prices for wireless earbuds and show gaps
Got it. I'll scrape your store and competitor pages for matching SKUs, then generate a price comparison report.
Tool Call
python3 << 'EOF'
import httpx, os

api_url = "https://api.webscraperapi.ai/v2"
api_key = os.environ["WEBSCRAPERAPI_API_KEY"]
headers = {"Authorization": f"Bearer {api_key}"}

pairs = [
  {
    "sku": "AirPods Pro 2",
    "my_store": "https://my-store.com/products/airpods-pro-2",
    "competitor": "https://competitor-store.com/products/airpods-pro-2",
  },
  {
    "sku": "Sony WF-1000XM5",
    "my_store": "https://my-store.com/products/sony-wf-1000xm5",
    "competitor": "https://competitor-store.com/products/sony-wf-1000xm5",
  },
]

for pair in pairs:
  my_price = httpx.get(
    f"{api_url}/scrape",
    params={"url": pair["my_store"], "prompt": "Extract current price", "llm": "gpt-4o-mini"},
    headers=headers,
  ).json()

  competitor_price = httpx.get(
    f"{api_url}/scrape",
    params={"url": pair["competitor"], "prompt": "Extract current price", "llm": "gpt-4o-mini"},
    headers=headers,
  ).json()
EOF
Price comparison (my store vs competitor):

  AirPods Pro 2
    my-store.com:         $249
    competitor-store.com: $229  (we are +$20)

  Sony WF-1000XM5
    my-store.com:         $259
    competitor-store.com: $279  (we are -$20)

Summary:
  • 1 SKU priced above competitor
  • 1 SKU priced below competitor
  • Report saved: reports/price-gap-wireless-earbuds.md
Done — I compared your store against the competitor and flagged SKU-level price gaps. Want me to monitor this daily and alert you when gaps exceed $10?

Agent monitors competitor pricing

Tell your agent to compare your store against competitor stores. It scrapes both sites, matches SKUs, and returns a clear price-gap report you can use for competitor monitoring.

Claude Code in action

In Claude Code, the agent can run real growth workflows, inspect output, and iterate until the task is complete, while calling webscraperapi endpoints under the hood.

Claude Code
 ▐▛███▜▌
▝▜█████▛▘
  ▘▘ ▝▝
Claude Code v2.1.59Opus 4.6 · Claude Max~/Documents/2026 Tax documents
/clear
└ (no content)
audit our pricing page against 3 competitors and draft fixes
webscraperapi - google_search (MCP)
webscraperapi - scrape (MCP)
Write (reports/pricing-audit.md)
Combobulating... (thinking)
Done. Found 4 conversion gaps and drafted exact copy fixes in reports/pricing-audit.md.
esc to interrupt

MCP: just add the server

Add one config block and every tool is available to your AI agent instantly.

MCP Tool Callwebscraperapi_scrape
1.2s
Parameters
url: https://news.ycombinator.com
output: markdown
Response
# Hacker News
1. Show HN: Open-source AI agent framework
2. Why Rust is taking over systems programming
3. The future of web scraping with LLMs...

Any HTTP client works

Plain curl, Python requests, fetch. If it speaks HTTP, it works.

bash
$ curl -s "https://api.webscraperapi.ai/v2/scrape?url=https://example.com&output=markdown" \ -H "Authorization: Bearer wsa_sk_..." | head -20 # Example Domain This domain is for use in illustrative examples in documents...
0M+
Requests processed
99.9%
Uptime
< 2s
p95 response time
02

You Handle the AI.
We Handle the Web.

Anti-bot systems, JavaScript rendering, captchas, rotating proxies, rate limits. That is a full engineering project. We handle all of it. Your agent sends a request, gets clean data back.

Never get blocked

Anti-bot bypass, proxy rotation, captcha solving. Invisible to you.

HTTP 403 Forbidden
HTTP 200 OK

Bloated context

~68KB sent

Lean context

~4KB sent

Same system and user prompts, smaller web payload, better output.

Keep the agent context window light

Agent context is layered: system prompt, user prompt, app instructions, and web data. Keep the web layer small with clean markdown, JSON, or CSV for better reasoning and lower cost.

Never maintain scrapers

Sites change. We adapt. Your API call stays the same.

soup.select_one(
  "div.price-box__old > span.now"
)
el = driver.find_element(
  By.XPATH,
  '//div[@class="product-info"]'
  '/span[contains(@class,"sale")]'
)
price = re.search(
  r'\$[\d,]+\.\d{2}',
  el.text
)
03

Try It Yourself

Pick an endpoint, pick a language. One call does it all.

curl -G "https://api.webscraperapi.ai/v2/scrape" \
  -H "Authorization: Bearer YOUR_KEY" \
  -d "url=https://example.com" \
  -d "output=markdown" \
  -d "render_js=true"
Response
# Example Domain

This domain is for use in illustrative
examples in documents. You may use this
domain in literature without prior
coordination or asking for permission.

[More information...](https://www.iana.org/
domains/example)
04

What Your Agents Can Do

Prompt: “Find 10 Series A fintech companies using React

Agent actions

  1. 1. Search Google for "Series A fintech React"
  2. 2. Scrape top 10 company sites
  3. 3. Extract founder names and emails
  4. 4. Verify email deliverability

Result: 10 qualified leads with verified contacts

🦞
Clawie
OpenClaw Agent

Find 10 AI startups in Berlin and get verified founder emails

9:12 AM

🔧webscraperapi_google_search
query: "Berlin AI startups" output: "json"
10 startup sites found
🔧webscraperapi_scrape
url: "https://example-startup.com" prompt: "Extract founders + emails"
Found 2 founders, 2 candidate emails
🔧webscraperapi_verify_email_address
email_address: "founder@example-startup.com"
{ "deliverable": true, "catch_all": false }

Done. I found 10 startups and verified 14 founder emails. Want me to export this as CSV and sync to your CRM?

9:13 AM

Find 10 AI startups in Berlin and get verified founder emails
05

Your AI Already Knows
How to Use It

📄llms.txt / llms-full.txt

Machine-readable docs indexes your agent can fetch and parse on first contact.

🧩SKILL.md

Drop-in setup instructions agents can follow without human help.

🔌MCP

One command to connect. Tools are discovered automatically via claude mcp add.

📐/docs · /openapi.json

Interactive docs and full OpenAPI schema. Every endpoint, every param, typed.

06

One API Instead of Five

Scraping, search, extraction, email verification, and tech detection. One API key. One bill.

FeaturewebscraperapiFirecrawlJina ReaderScrapingBeeDIY
Scrape any URL to markdownPartialDIY
JavaScript renderingDIY
Anti-bot bypass + captchas
LLM-powered extractionPartial
Dedicated Google Search APIPartialPartial
Google Shopping API
Email verification
Tech stack detection
MCP tools (Claude/Cursor)
Maintenance requiredNoneNoneNoneNoneConstant
Free tier500 credits500 creditslimited1,000 trial callsn/a
Starting paid price$19/mo$16/mousage$49/mo$$$+

Based on public docs and pricing pages (Feb 2026). Feature scope can change — verify current docs before purchasing.

07

All Features. Every Plan.

Just pick your credits. No feature gates. Need more? Top up anytime.

🌐 Scrape

  • JS rendering
  • LLM extraction
  • CSS selectors

🔍 Google Search

  • Locale & geo targeting
  • Pagination

🛒 Google Shopping

  • Currency & locale
  • LLM extraction

📢 Google Ads

  • Paid ad results
  • Domain filtering

✉️ Email Verification

  • Deliverability check

🛡️ Tech Detection

  • Stack identification

📊 SERP Frequencies

  • Competitive analysis

🔌 MCP Server

  • OAuth authentication
  • Auto tool discovery

📄 Output Formats

  • Markdown, JSON, HTML, CSV

Free

$0/mo

500 credits/mo

No credit card required.

  • All features included
  • 5 concurrent requests
  • Best-effort support

webscraperapi plus

$19/mo

5,000 credits/mo

  • All features included
  • 10 concurrent requests
  • Standard support
Most Popular

webscraperapi pro

$99/mo

50,000 credits/mo

  • All features included
  • 50 concurrent requests
  • Standard support
Save $91

webscraperapi max

$199/mo

150,000 credits/mo

  • All features included
  • 200 concurrent requests
  • Priority support
Save $371
08

Common Questions

Give Your Agents
the Whole Web.

500 free credits. All endpoints. No credit card.
Your agents start pulling live web data in under a minute.

No credit card required
Up and running withOpenClawin 2 minutes