FlowCay logo
FlowCay
Bright Data MCP Claude Skill Deep Research

Bright Data MCP Claude Skill Deep Research

Open-source deep research Skill combining Claude Code, MCP and Bright Data for multi-platform e-commerce channel research.

53

Install command

npx skills add https://github.com/liangdabiao/Bright-Data-MCP-Claude-Skill-deep-research
SkillOpen SourceAll Platforms
Added on Popularity 4

Details

Overview

Bright Data MCP Claude Skill Deep Research (by liangdabiao, open source, released Jan 2026) combines three hot technologies — Claude Code, the Model Context Protocol (MCP) and Agent Skills — to run deep multi-channel e-commerce / market research automatically. It pairs Bright Data's official MCP server (which provides the network/data tools) with a custom research-brightdata Claude Skill (which encodes the author's research methodology), plus a DuckDuckGo search skill. The result: type something like "deep research the 3D printer market on Amazon" and Claude orchestrates search, scraping, structured extraction and report generation end-to-end.

Key Features

  • Multi-phase research workflow: query/scope analysis → source discovery via search_engine → content collection → AI-powered structured extraction → validation & normalization → synthesized report, JSON dataset or Markdown summary.
  • Three collection modes: Quick (scrape_as_markdown per URL), Standard (scrape_batch, 10 URLs at once), and Deep (browser automation — navigate, links, click — for JavaScript-rendered or interactive pages).
  • Anti-bot protection: Bright Data's proxy network and anti-scraping technology keep collection stable against bot detection, CAPTCHAs and geo restrictions.
  • Structured extraction: AI extract tool pulls fields like product name, price, rating, reviews and seller into normalized tables.
  • Multi-dimensional analysis: market size and trends, brand competition, pricing strategy, consumer behavior, supply chain, tech trends.
  • Market research example output: shipped Amazon 3D-printer deep-research report (Jan 24, 2026) covering global market size ($15.4B in 2024), consumer-segment data (Q1 2025: 1M+ entry-level printers shipped, 95% made by Chinese manufacturers), and more.
  • Simple setup: configure your Bright Data API key in .mcp.json, then invoke via natural language in Claude Code — e.g. "deep research the 3D printer market on Amazon" or "deep research anime figure brand competition on eBay".
  • Complementary official tooling: Bright Data also publishes its own brightdata/skills repo (agent-onboarding, search, scrape, data-feeds, bright-data-mcp with 60+ MCP tools, discover-api, live-research, rag-pipeline, competitive-intel), which the community projects build on.

How to Use

  1. Use a Skill-compatible agent such as Claude Code, Cursor, or Codex.
  2. Install from the terminal (Node.js required):
npx skills add https://github.com/liangdabiao/Bright-Data-MCP-Claude-Skill-deep-research

Or paste the install prompt from this tool page into your agent and ask it to install for you. 3. Restart / reload the agent so the new Skill is loaded. 4. Invoke the Skill with its slash commands or by asking the agent to run the workflow described in Overview / Key Features. 5. Review outputs before publishing — treat model-generated copy, images, and data as drafts.

Pricing

  • Skill itself: free and open source on GitHub (MIT-style community project) — no license cost for the workflow.
  • Data/network layer: requires a Bright Data account with API credits (Bright Data is a paid global proxy / web-scraping / datasets platform). Costs depend on your scraping volume, IP/bandwidth and MCP API usage; Bright Data offers a free tier / trial for new users. The official brightdata/skills repo notes its competitive-intel skill replaces ~$15K+/yr enterprise CI tools "at pennies per analysis".

Best For

Cross-border e-commerce sellers, researchers and analysts who want a repeatable, methodology-driven deep-research workflow across platforms (Amazon, eBay, Shopify, etc.) — market sizing, competitor analysis, price monitoring, review mining and trend spotting — without hand-writing scrapers or paying for boutique research agencies.

FAQ

What do I need to run it?

Claude Code (or another SKILL.md-based agent), a Bright Data account/API key configured in .mcp.json, and Python/Node tooling for scripts. The Skill files are cloned from the GitHub repo.

Does it work with JavaScript-heavy sites?

Yes — "Deep" mode uses browser automation (navigate/click/links) for dynamic or interactive pages, in addition to quick batch scraping.

How is MCP different from a Skill here?

The MCP server (by the data provider) exposes Bright Data's own APIs to the agent — code nobody should re-implement; the Skill is the author's private research methodology, which only the user can define. The project combines both.

Is it only for Amazon?

No — the workflow targets any e-commerce channel or platform; the README gives eBay brand-competition research as another example, and official Bright Data skills cover Amazon, Walmart, eBay, Best Buy, Etsy, etc.

How much does it cost to run?

The Skill is free. You pay Bright Data for API/network usage (proxy bandwidth and API calls). For light research volumes that's cents per analysis; heavy large-scale collection scales with volume.