Bright Data
Also known as: Bright Data Web MCP, Luminati Networks, Web Unlocker, Scraping Browser
Web data infrastructure for agents, with an MCP server of 60-plus tools, unblocking and SERP APIs, managed cloud browsers, 1,300-plus pre-built scrapers and a 400-million-IP proxy network.
Bright Data is web data infrastructure for AI agents and data teams. Agents reach it through an MCP server with more than 60 tools, from web search and page unlocking to structured extractors for sites such as Amazon, LinkedIn and Instagram and a set of browser tools that navigate, click and type, grouped into 11 use-case bundles.
Behind the tools sit the Web Unlocker API, which fetches any public page past CAPTCHAs and anti-bot systems, the SERP API, a Browser API that runs Playwright, Puppeteer or Selenium code in managed cloud browsers, 1,300-plus pre-built scrapers, and Scraper Studio, which builds a custom scraper from a prompt and repairs it when the site changes, holding each AI fix for the user's approval.
Its proxy networks cover more than 400 million residential IPs across 195-plus countries, with residential access limited to KYC-verified business accounts, and it sells 350-plus ready-made datasets. Results can be delivered to a webhook or straight into S3, Google Cloud Storage, Azure, Snowflake or SFTP. Bright Data publishes ISO 27001, 27017 and 27018 certificates, a SOC 2 Type II report under NDA and a downloadable SOC 3.
Pricing is usage-based, with 5,000 free credits a month across the Unlocker, SERP, Scraper and Studio products (and the Browser API from September 2026), then Web Unlocker and SERP from $1 per 1,000 requests, scrapers from $0.75 per 1,000 records, the Browser API from $5 per GB and residential proxies from $2.50 per GB.
Bright Data, formerly Luminati Networks, has been majority owned by the private equity firm EMK Capital since 2017.
Vendor details
Canonical URL
https://brightdata.com
Category
Agent infrastructure
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
A remote and local MCP server with 60-plus tools in 11 use-case groups, REST APIs for every product, Python and JavaScript SDKs, a CLI and Agent Skills. Alongside them come 1,300-plus pre-built scrapers, Scraper Studio for custom scrapers, and result delivery to webhooks, S3, Google Cloud Storage, Azure, Snowflake or SFTP.
In practice
Your agent needs the live web and keeps getting blocked. The Web MCP server gives Claude, Cursor or Windsurf unblocked search, navigation and extraction, with 5,000 requests a month free to start.
Your legal team asks whether collecting this data is defensible. Bright Data publishes an ethics and compliance report and vets every customer through KYC before granting access, with residential proxies limited to KYC-verified business accounts.
Procurement needs certifications, not assurances. SOC 2 Type II with a downloadable SOC 3, ISO 27001:2022, 27017, 27018 and 27701 and a CSA STAR listing are all published.
Sources & related URLs
Agentic Index coverage score
8.0 / 14 capabilities · 57%
| Integrations & Tool Calling | Partial |
|---|---|
|
Agents get web data as tools. The MCP server's more than 60 tools cover search, page unlocking and structured extractors for sites such as Amazon, LinkedIn and Instagram. The Web Scraper API offers more than 1,300 prebuilt scrapers, and Scraper Studio builds custom scrapers from a prompt. All of these read public web data. They take no actions in the customer's own systems and hold no credentials for individual users. Sourcedocs.brightdata.com/products/mcp-server/toolsread 2026-09-22 |
|
| Workflow Orchestration | Not documented |
|
Each job Bright Data runs is a collection job. A scraper is triggered, runs on Bright Data's infrastructure and delivers results. Customers cannot sequence, branch or route multiple steps into a workflow. Bright Data's tutorials chain SERP and scraper calls in the customer's own code or in GitHub Actions. Sourcedocs.brightdata.com/llms.txtread 2026-09-22 |
|
| Knowledge Grounding & RAG | Not documented |
|
What Bright Data supplies is public web data, ready made datasets and a people and company search index that it maintains. None of these draws on the customer's own documents or data. There is no ingestion, retrieval or RAG layer over company data. Sourcedocs.brightdata.com/llms.txtread 2026-09-22 |
|
| Human Oversight & Guardrails | Full |
|
A person approves AI changes before they commit. When Scraper Studio's AI self-healing rewrites a broken scraper, the heal stops by default at an approval gate with status awaiting_approval and a preview of the result. The user then approves the fix, which commits it, or rejects it. Spend limits for each zone and zone permissions on API keys add further constraints. Sourcedocs.brightdata.com/products/cli/commandsread 2026-09-22 |
|
| Security, Identity & Governance | Full |
|
ISO/IEC 27001:2022, 27017 and 27018 certificates cover Bright Data's web data platform, including the MCP server and Browser API, and each certificate's number and validity dates are public. Bright Data shares a SOC 2 Type II report under NDA and offers a publicly downloadable SOC 3 from a Deloitte network firm. It also has a CSA STAR listing and GDPR and CCPA programs. API keys are bound to zones with their own permissions and admin rights, and residential proxies require a KYC-verified business account. Sourcedocs.brightdata.com/general/security/security-overviewread 2026-09-22 |
|
| Observability & Auditability | Partial |
|
The control panel records usage. Its Cost Explorer and zone statistics break usage down by product, zone and hour. The Proxy Manager, which customers host themselves, can export request logs to a webhook, Datadog, Logz.io or S3. There is no step by step record of what an agent asked for and received through the MCP server or the APIs. Sourcedocs.brightdata.com/products/proxy-manager/faqsread 2026-09-22 |
|
| Memory & State Persistence | Not documented |
|
Job snapshots hold collection results for download, and sessions are held for browsing. No memory layer keeps an agent's context across runs. Sourcedocs.brightdata.com/llms.txtread 2026-09-22 |
|
| Deployment & Data Residency | Partial |
|
Collection runs on Bright Data's network. Results can land in the customer's own infrastructure, such as S3, Google Cloud Storage, Azure, Snowflake, SFTP or a webhook. The MCP server and the open source Proxy Manager can run locally. Customers cannot choose the region where Bright Data processes and stores data, and there is no self hosted collection platform. Sourcedocs.brightdata.com/products/scrapers/scrapers-library/data-deliveryread 2026-09-22 |
|
| Prebuilt Agents, Templates & Packs | Partial |
|
Tools come packaged by use case. The MCP server offers 11 preconfigured groups, among them e-commerce, social media, browser automation, business intelligence, finance and research, and Bright Data publishes Agent Skills for coding agents. Bright Data also offers more than 1,300 prebuilt scrapers and sells more than 350 ready made datasets as data products. There is no catalog of ready made workflows or agents. Sourcedocs.brightdata.com/products/mcp-server/toolsread 2026-09-22 |
|
| Triggers & Channel Coverage | Full |
|
Collection can start and report without a person. Scraper Studio scrapers run through initiation endpoints, and a schedule is one of the initiation modes. When a job completes, scraper results are pushed to the customer's webhook for event driven pipelines, or to S3, Google Cloud Storage or Snowflake. Sourcedocs.brightdata.com/products/scrapers/scrapers-library/data-deliveryread 2026-09-22 |
|
| Model Flexibility & Routing | Full |
|
Any model the customer's agent runs can take Bright Data's web data. The MCP server works with any MCP client, and integrations with coding agents and agent frameworks each use the customer's own model. Bright Data has no model of its own for agents, so the customer chooses the model. Sourcedocs.brightdata.com/products/mcp-server/overviewread 2026-09-22 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
REST APIs cover every product, from Web Scraper, Scraper Studio, Web Unlocker and SERP to account management and the dataset marketplace. Python and JavaScript SDKs, a CLI and Agent Skills sit alongside them, as do a remote and local MCP server with OAuth and a protocol that lets an agent register an account without a signup form. Sourcedocs.brightdata.com/llms.txtread 2026-09-22 |
|
| Testing, Debugging & Optimization | Not documented |
|
Error codes, previews of self-healed scrapers and fill rate statistics for datasets check Bright Data's own scrapers and data. Customers have no way to test or score their own agents. Sourcedocs.brightdata.com/llms.txtread 2026-09-22 |
|
| Browser & Computer Use | Full |
|
The Browser API runs the customer's Playwright, Puppeteer or Selenium code in fully managed cloud browsers built for user-like interactions, and Bright Data's MCP server gives an agent browser tools to navigate, click, type, scroll, wait for elements and take screenshots. Sourcedocs.brightdata.com/products/scraping-browser/introductionread 2026-09-22 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
Version 2.12.0 of the Bright Data MCP server adds six Web Scraper API tools that let an agent search a self updating catalog of more than 1,200 prebuilt scrapers, run one, and poll for results when a run takes longer than 45 seconds. The same release adds a content gate that asks the user, through MCP elicitation, before returning scraped content that looks like instructions to change the MCP configuration.
Bears on: Human approval / guardrails
View sourceBright Data released an integration with Paperclip, an open-source AI orchestration platform for managing teams of agents. The integration allows Paperclip agents to access the web using Bright Data's infrastructure via the Model Context Protocol (MCP) and Agent Skills.
Bears on: Integrations
View sourceBright Data introduced an integration with Grok Build, allowing the terminal-based AI coding agent to access the live web. The integration utilizes the Bright Data Web MCP to enable web search, scraping, discovery, and browser automation directly within Grok Build.
Bears on: Integrations
View sourcePricing
Free 5,000 credits/mo · Web Unlocker and SERP from $1/1K requests · scrapers from $0.75/1K records · Browser API from $5/GB
gigabytes transferred, requests, or successful results depending on product
Included quota
Free tier: 5,000 credits a month shared across Web Unlocker, SERP API, Web Scraper API and Scraper Studio (one credit per request or record, one per page load in Studio), and the Browser API at 5 credits per MB from 1 September 2026. Proxies are excluded.
What is public
Starting prices for every product, the free tier and its credit rules are published; volume discounts are negotiated.
Billing mechanics
Prepaid wallet with a recurring free monthly credit allowance, then pay per use in each product's own unit: requests, records, gigabytes or IPs.
Cost watchouts
Each product meters on its own unit (requests, records, gigabytes or IPs), so an agent that mixes products draws on several meters at once. The prepaid wallet stops when funds run out rather than billing a surprise overage.
Variable cost rationale
Cost tracks bandwidth and request volume, both of which an autonomous agent controls rather than the buyer, and two of the four billing modes charge for failed attempts, so a poorly behaved agent hitting defended targets can consume budget without producing results.
Additional watchouts
Residential proxies require a KYC-verified business account.
Overage / add-ons
Prepaid wallet: usage draws down deposited funds and a free account stops when credits run out; no surprise bill.
Sales call required
Mixed (some tiers require a call)
Free / trial
5,000 free credits a month, no card; proxies get a separate one-time trial credit
Lowest paid plan
Pay per use from $0.75 per 1,000 scraper records or $1 per 1,000 Web Unlocker or SERP requests
Commercial notes
Self-serve with a free monthly allowance and agent-driven account registration; enterprise and volume terms are negotiated.
Key ambiguities
Published figures are starting prices; volume and enterprise rates are negotiated.
Missing data
Negotiated volume and enterprise rates.
Related vendors
- AgentOps — Agent observability and debugging platform: open source SDKs trace…
- AgentPhone — AgentPhone gives the AI agents a team builds their own US or…
- Agno — Python agent framework and AgentOS runtime (formerly Phidata) for…
- AIsa — Resource and payment gateway for AI agents, with one key to 110+…
- Albato — Albato is a no code integration platform that sells an embedded,…
- AlphaBitCore — AI control plane for regulated financial firms: one gateway enforces…
Alternatives to Bright Data
The closest documented capability profiles to Bright Data among agent infrastructure platforms tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Firecrawl8.5 / 14Fuller documented coverage on Observability & Auditability and Deployment & Data Residency
- Exa6.5 / 14Adds documented Memory & State Persistence
- boxd9.0 / 14Adds documented Workflow Orchestration and Memory & State Persistence
- Daytona8.0 / 14Adds documented Memory & State Persistence
- E2B8.0 / 14Adds documented Memory & State Persistence
- F5 AI Guardrails9.0 / 14Adds documented Testing, Debugging & Optimization
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded