# String

> String is a web scraping API for reliable web data. Fetch any page on the web - anti-bot handling built in, and you only pay when content comes back.

This file is generated from the site's source content. It is a curated index for AI assistants, not a crawler permissions file. For crawler permissions, see `/robots.txt`.

The homepage, every product page (Web Access API, Pricing, Products, Composer, Bespoke Web Datasets, Finance, About), the blog index, every blog post, the comparisons index, every comparison page, the answers index, and every answer page are also available as Markdown: request the page with an `Accept: text/markdown` header, or append `.md` to the URL (the homepage is `/index.md`).

## Core Pages

- [Home](https://usestring.ai/): Primary overview of the site.
- [Products](https://usestring.ai/products): The three String products at a glance.
- [Web Access API](https://usestring.ai/web-access): Self-serve API to reliably fetch any page on the web, with the full Web Data Frontier Benchmark.
- [Benchmark](https://usestring.ai/benchmark): The Web Data Frontier Benchmark, the current leaderboard and the methodology, rerun regularly.
- [Composer](https://usestring.ai/composer): Describe a site and receive a structured, scheduled feed. Limited beta access is open.
- [Bespoke Web Datasets](https://usestring.ai/managed-services): We run the data pipelines for you, with an SLA.
- [Finance](https://usestring.ai/finance): Managed web data signals for investment teams building an independent earnings view.
- [Pricing](https://usestring.ai/pricing): Starter $20/month or Growth $100/month, plus usage billed per 1,000 successful requests (Fetch · standard proxy: $0.30 / 1k on Starter, $0.20 / 1k on Growth; browser and premium-proxy rates on the page). You only pay for data successfully delivered.
- [About](https://usestring.ai/about): Built in New York City.
- [Careers](https://usestring.ai/careers): Open roles at String.
- [Book a demo](https://usestring.ai/book-demo): Tell us what web data you need and we route you to the right team.
- [Blog](https://usestring.ai/blog): Notes from the String team on web access, benchmarks, and building for agents.
- [Comparisons](https://usestring.ai/comparisons): Side-by-side comparisons of web access and scraping APIs: String next to Firecrawl, Bright Data, Apify, and others, one page per vendor, kept current by the String team.
- [Answers](https://usestring.ai/answers): Direct answers to the questions buyers ask about web scraping and web access: legality, compliance, cost, anti-bot handling, and agent data pipelines. One question per page, every number sourced.

## Legal

- [Terms of service](https://usestring.ai/terms)
- [Privacy policy](https://usestring.ai/privacy)

## Trust and Security

- [Trust center](https://usestring.ai/trust): The security, confidentiality, and availability practices that protect String's services. String AI is pursuing a SOC 2 Type I examination; no audit report has been issued, and String AI does not claim to be SOC 2 certified, compliant, or audited. EU and UK GDPR Article 27 representatives are appointed under certificate EV103335.
- [Security](https://usestring.ai/security): How to report a vulnerability, and the security contact.
- [Sub-processors](https://usestring.ai/subprocessors): The third parties that support String's services.

## Blog Posts

- [Web Data Frontier Benchmark: the August 2026 run](https://usestring.ai/blog/web-data-frontier-benchmark-august-2026): Second official run of the open benchmark: 99 real-world targets, 15 providers, 7,425 live requests. Full leaderboard and what moved since July.
- [How to build your own GEO engine](https://usestring.ai/blog/growth-diaries-geo-engine): We researched the GEO tools on the market, then built our own tracker and improvement engine on the Web Access API. 200 questions, 9 surfaces, a Monday review, and a daily draft queue - open sourced so you can build yours.
- [String's Web Access API is now on Apify: 30 Actors](https://usestring.ai/blog/apify-actors): 30 single-site scrapers on the Apify Store, each one a thin wrapper on the String Web Access API: job boards and ATS portals, e-commerce and retail, real estate, reviews, finance and ticketing, and public profiles.
- [The web scraping industry has a benchmark problem](https://usestring.ai/blog/web-scraping-benchmark-problem): Every vendor claims a 98-100% success rate, and they can't all be right. So we open-sourced a benchmark anyone can run.
- [Introducing the String Web Access API](https://usestring.ai/blog/introducing-web-access-api): State-of-the-art web access for humans and agents
- [Our Web Access Manifesto: Information wants to be free](https://usestring.ai/blog/web-access-manifesto): We're building String to help the world access information on the web at scale

## Comparisons

- [Firecrawl Alternatives 2026: Free, Open Source and Hosted](https://usestring.ai/comparisons/firecrawl-alternatives): Firecrawl and six alternatives on the September 2026 benchmark of 100 bot-protected sites, plus six free and open-source libraries, the enterprise picks, verified pricing at $100 a month, and where Firecrawl is still the right choice. Updated September 24, 2026.
- [String vs Context.dev (2026): 100 protected sites](https://usestring.ai/comparisons/string-vs-context-dev): String vs Context.dev on 100 bot-protected sites in September 2026: 97.0% against 72.0%, latency, flat-credit pricing, and what changed between runs. Updated September 23, 2026.
- [String vs Decodo (2026): 100 protected sites](https://usestring.ai/comparisons/string-vs-decodo): String vs Decodo on 100 bot-protected sites in the September 2026 benchmark: 97.0% against 50.6%, strongest-mode testing, premium-plus-JS pricing, and what changed between runs. Updated September 23, 2026.
- [String vs Nimble (2026): 100 protected sites](https://usestring.ai/comparisons/string-vs-nimble): String vs Nimble on 100 bot-protected sites in September 2026: 97.0% against 68.6%, pricing, product breadth, and where Nimble wins. Updated September 23, 2026.
- [String vs ZenRows (2026): 100 sites compared](https://usestring.ai/comparisons/string-vs-zenrows): String vs ZenRows on 100 real sites in September 2026: success, latency, credit multipliers, pay-per-success billing, browser tools, and fit. Updated September 23, 2026.
- [String vs Zyte (2026): 100 sites compared](https://usestring.ai/comparisons/string-vs-zyte): String vs Zyte on an open benchmark of 100 real sites in September 2026: success rate, latency, tiered pricing, HTTP and browser paths, and fit by job. Updated September 23, 2026.
- [Web Scraping API Pricing in 2026: 16 Providers Compared](https://usestring.ai/comparisons/web-scraping-api-pricing): Standard vs premium pricing for 16 web scraping APIs, credit multipliers, which failures still bill, and the rate each provider charges for the configuration it ran on a 100-site benchmark. Updated September 23, 2026.
- [Best E-commerce Scraping APIs in 2026: 8 APIs Compared](https://usestring.ai/comparisons/best-ecommerce-scraping-apis): Eight e-commerce scraping APIs measured on the 43 retail, fashion, marketplace and grocery sites in the September 16, 2026 Web Data Frontier Benchmark. String, Scrapfly, ScraperAPI, Firecrawl, Oxylabs, Apify, ScrapingBee and Bright Data, with per-vertical success rates, verified pricing, and the sites that still open for nobody. Updated September 22, 2026.
- [Best Web Scraping APIs in 2026: 16 Tools Compared](https://usestring.ai/comparisons/best-web-scraping-apis): Sixteen web scraping APIs measured against the same 100 bot-protected sites, five attempts each, plus verified pricing at the $100-a-month tier, free-tier sizes, and the billing rule each vendor publishes. Success rates, latency, anti-bot results, and a pick for each job. Updated September 22, 2026.
- [String vs ScraperAPI (2026): 100 sites compared](https://usestring.ai/comparisons/string-vs-scraperapi): String vs ScraperAPI on 100 real sites in the September 2026 benchmark: success, latency, credit multipliers, billing, MCP support, and which product fits which workload. Updated September 22, 2026.
- [String vs Scrapfly (2026): 100 sites compared](https://usestring.ai/comparisons/string-vs-scrapfly): String and Scrapfly compared on 100 real sites in the September 2026 benchmark: success rate, latency, anti-bot coverage, pricing, product breadth, and where each wins. Updated September 22, 2026.
- [String vs ScrapingBee (2026): 100 sites compared](https://usestring.ai/comparisons/string-vs-scrapingbee): String and ScrapingBee compared on 100 real sites in the September 2026 benchmark: coverage, latency, pricing, Auto Mode, MCP, extraction, and where each product fits. Updated September 22, 2026.
- [Best Apify Alternatives in 2026: 8 Tools Compared](https://usestring.ai/comparisons/apify-alternatives): Seven Apify alternatives on open benchmark data, compared on pricing model and output format, plus the jobs where a prebuilt Actor still wins. Updated September 22, 2026.
- [Best Scrape.do Alternatives in 2026: 8 Tools Compared](https://usestring.ai/comparisons/scrape-do-alternatives): Seven Scrape.do alternatives compared on a 99-site web scraping benchmark and on pricing checked against each vendor's own pages: success rate, response times, cost near $100 a month, anti-bot results and MCP support. String, Scrapfly, Context.dev, Firecrawl, Bright Data, Oxylabs and Zyte, plus what Scrape.do's own price list says about the sites it finds hard. Updated September 18, 2026.
- [ScraperAPI vs ZenRows vs ScrapingBee: which to use in 2026](https://usestring.ai/comparisons/scraperapi-vs-zenrows-vs-scrapingbee): A three-way comparison of ScraperAPI, ZenRows and ScrapingBee built from a 99-site benchmark run and each vendor's own pricing docs. Includes the targets where only one of the three returns a page, what each costs at its strongest setting, and the configuration caveat that applies to two of them. Updated September 18, 2026.
- [Best ScraperAPI Alternatives in 2026: 6 Tools Compared](https://usestring.ai/comparisons/scraperapi-alternatives): Five ScraperAPI alternatives on a 99-site benchmark and verified pricing: String, Scrapfly, Firecrawl, Bright Data, Zyte. Success, latency, cost. Updated September 18, 2026.
- [Best ZenRows Alternatives in 2026: 7 Tools Compared](https://usestring.ai/comparisons/zenrows-alternatives): Six ZenRows alternatives on a 99-site scraping benchmark: String 97.0% against ZenRows 34.3%, verified pricing, and what $100 a month buys. Updated September 18, 2026.
- [Best Web Scraping Tools in 2026: 11 Tools Compared](https://usestring.ai/comparisons/best-web-scraping-tools): Eleven web scraping tools compared by the job they do: hosted scraping APIs measured on a 99-site benchmark, a scraper marketplace, open-source frameworks and a no-code app. String, Scrapfly, Context.dev, Firecrawl, Bright Data, Oxylabs, Apify, Crawl4AI, Scrapy, Playwright and Octoparse, with verified pricing and how many tools a list of hard sites really needs. Updated September 13, 2026.
- [Best Nimble Alternatives in 2026: 8 Tools Compared](https://usestring.ai/comparisons/nimble-alternatives): Seven Nimble alternatives compared on a 99-site web scraping benchmark and on pricing verified against each vendor's own pages: success rate, latency, cost at $100 a month, results behind Akamai and Cloudflare, and MCP support. String, Scrapfly, Context.dev, Firecrawl, Bright Data, Oxylabs and Decodo, plus the case for staying on Nimble. Updated September 11, 2026.
- [String vs Bright Data (2026): 99 sites compared](https://usestring.ai/comparisons/string-vs-bright-data): String vs Bright Data on an open benchmark of 99 real sites: success rate, latency, pricing, product breadth, and which one fits which job. Updated September 10, 2026.
- [String vs Oxylabs (2026): 99 sites compared](https://usestring.ai/comparisons/string-vs-oxylabs): String vs Oxylabs on an open benchmark of 99 real sites: success rate, latency, pricing, proxy products, and which one fits which job. Updated September 10, 2026.
- [String vs Firecrawl (2026): compared on 99 real-world sites](https://usestring.ai/comparisons/string-vs-firecrawl): String vs Firecrawl on an open benchmark of 99 real sites: success rate, latency score, price per usable page, and which to pick for which job. Updated September 10, 2026.
- [Best Bright Data Alternatives in 2026: 7 Tools Compared](https://usestring.ai/comparisons/bright-data-alternatives): Six Bright Data alternatives on a 99-site scraping benchmark: String 97.0% against Bright Data 78.0%, verified pricing, and what $100 a month buys. Updated August 27, 2026.
- [Best Oxylabs Alternatives in 2026: 7 Tools Compared](https://usestring.ai/comparisons/oxylabs-alternatives): Six Oxylabs alternatives on a 99-site scraping benchmark: String 97.0% against Oxylabs 76.8%, verified pricing, and what $100 a month buys. Updated August 27, 2026.
- [Best Zyte Alternatives in 2026: 7 Tools Compared](https://usestring.ai/comparisons/zyte-alternatives): Six Zyte alternatives on a 99-site benchmark and verified pricing: String, Scrapfly, Context.dev, Firecrawl, Bright Data, Oxylabs. Success, latency, cost. Updated August 26, 2026.

## Answers

- [Does robots.txt legally prevent web scraping?](https://usestring.ai/answers/does-robots-txt-legally-prevent-scraping): No. robots.txt is a voluntary protocol, not law, and no US court has held that ignoring it is by itself unlawful. It still changes your legal position. Updated August 29, 2026.
- [What is the hiQ vs LinkedIn ruling and does it still apply?](https://usestring.ai/answers/hiq-v-linkedin-ruling): hiQ won the CFAA fight and still lost the case. The Ninth Circuit protected public scraping; a contract claim ended the company. Both halves still apply. Updated August 29, 2026.
- [Is it legal to scrape competitor prices?](https://usestring.ai/answers/is-it-legal-to-scrape-competitor-prices): Prices are facts, and facts are not copyrightable. Public price collection is the best-supported case in scraping law. The constraints are contract and antitrust. Updated August 29, 2026.
- [Is web scraping legal?](https://usestring.ai/answers/is-web-scraping-legal): Scraping public data is generally lawful in the US, and the real risk sits in contract and privacy law rather than hacking statutes. What the cases actually held. Updated August 29, 2026.

## Integrations

- [MCP server](https://portal.usestring.ai/docs/mcp/remote): Hosted endpoint https://mcp.usestring.ai/v1/mcp (Streamable HTTP; Bearer API key, or the key in the path). Add it as a custom connector in Claude.ai or Perplexity, or as a config entry in Cursor, ChatGPT, and any other MCP client.

## Discovery

- [Sitemap](https://usestring.ai/sitemap.xml): Complete search-engine sitemap, if available.
- [Robots policy](https://usestring.ai/robots.txt): Crawler permissions and content signals for automated agents (Content-Signal: search=yes, ai-input=yes, ai-train=no; fetch, index, and cite freely, training use declined).