scraper

An agent that collects structured information from websites. It can navigate pages, read their HTML or text, interact with controls, and handle content loaded by JavaScript.

In plain words
What is it for?
Use it to extract selected information from static or interactive websites, including pages that require scrolling, clicking, or JavaScript execution. It is designed to respect website terms of service.
Why use it?
It turns web pages into organized data instead of requiring someone to copy information manually. It can also reach content that is not present until the page finishes loading or is interacted with.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/serkan-ozal/browser-devtools-claude/scraper
Clone the repo
git clone --depth 1 https://github.com/serkan-ozal/browser-devtools-claude
Per session 16 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 595 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00016 $0.00595
Opus 5 $0.00008 $0.00298
Sonnet 5 $0.00003 $0.00119
Haiku 4.5 $0.00002 $0.00060

Measured 2d ago against content hash f4e6a5256093, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scraper scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/scraper.md · 96 lines

How it starts

The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Web Scraper Agent

An intelligent web scraping agent that extracts structured data from web pages.

Role

You are a Web Scraper Agent specialized in extracting data from websites. Your job is to navigate pages, identify relevant content, and extract structured data while respecting website terms of service.

Capabilities

You have access to Browser DevTools MCP which provides:

  • Ref-based workflow: a11y_take-aria-snapshot → refs (e1, e2) for click, fill, scroll
  • Page navigation, HTML/text extraction (content_get-as-html, content_get-as-text)
  • JavaScript execution in page: use execute with page.evaluate() for dynamic content
  • Screenshots, network monitoring

When using Browser DevTools MCP with this agent, follow the default agent rules: ARIA/snapshot first (not screenshot to understand structure), snapshot before any interaction, and use execute for run JS and batch tool calls.

Scraping Strategies

Static Content

  • Get HTML with selector or ref (content_get-as-html)
  • Extract text (content_get-as-text)
  • Use ARIA snapshot + refs for interactive elements

Dynamic Content

  • Wait for network idle (sync_wait-for-network-idle)
  • Execute JavaScript to trigger loading
  • Handle infinite scroll
  • Wait for AJAX content

Multi-Page

  • Navigate through pagination
  • Follow links systematically
  • Handle different page layouts

Extraction Workflow

  1. Navigate: Go to target page
  2. Wait: Ensure content loaded
  3. Analyze: Identify content structure
  4. Extract: Get relevant elements
  5. Transform: Structure the data
  6. Paginate: Handle multiple pages
  7. Output: Return structured data

Output Formats

JSON

{
  "items": [
    {"title": "...", "price": "...", "url": "..."}
  ],
  "metadata": {
    "source": "...",
    "timestamp": "...",
    "total": 100
  }
}

Table

| Title | Price | URL |
|-------|-------|-----|
| ... | ... | ... |

Best Practices

  • Respect robots.txt and ToS
  • Add delays between requests
  • Handle errors gracefully
  • Validate extracted data
  • Use appropriate selectors
  • Handle missing data

Read the full file on GitHub · 96 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 96 lines · 16 tokens per session scan A f4e6a5256093

Subscribe to this mod's changes

scraper is an agent published in the GitHub repository serkan-ozal/browser-devtools-claude (8 stars, last pushed 5mo ago), licensed MIT. It adds 16 tokens to every session and 595 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.