Context.dev
by Context.dev
One web context API for AI agents: scrape, crawl, search, extract and enrich with a single key
Context.dev is a hosted web data API that gives AI agents and developers clean Markdown, screenshots, structured JSON, brand and company data, people enrichment, news and page monitoring through one API key. It is for engineering teams that want to replace a patchwork of scrapers, proxies and enrichment vendors behind their agents.
Context.dev is a Y Combinator (S26) company founded by Yahia Bakour that began as Brand.dev, an API for logos, colours and company descriptions used to pre-fill customer onboarding. Customers kept asking for screenshots, structured extraction and LLM-ready data, so the product widened into a general web context API and rebranded. Its versioned /v1 API covers scraping to HTML, Markdown, screenshots or CSS-parsed fields; URL mapping; crawls of up to 500 pages; ranked web search with domain and freshness filters; an Answers endpoint that researches a task and returns JSON in a supplied shape; brand retrieval and styleguide extraction; people and company search; live and historical news; file parsing to Markdown; and page, sitemap and extract monitors that fire signed webhooks. Every request is JavaScript-rendered on a custom browser stack with residential proxies, automatic escalation and retries, and failed or blocked requests are not billed. Official SDKs ship for TypeScript, Python, Ruby, Go and PHP under Apache-2.0, alongside an MCP server and a CLI. The company states SOC 2 Type 2 compliance and a 99.9% uptime SLA on Enterprise; it is hosted only. It launched on Hacker News in July 2026 (119 points, 86 comments), reported 400+ customers in August, and on October 19, 2026 announced a $9M round led by General Catalyst with Orange Collective and Robinhood Ventures, with the founder citing 800+ customers including Mintlify and daily.dev.
An engineering lead building agents or AI features that need live web data, who is currently stitching together a scraper, a proxy provider, a search API and an enrichment vendor.
One credit-metered API and SDK for scraping, search, extraction, enrichment and monitoring, with JS rendering and proxies included in the price.
At a Glance
- Category
- Developer Tools
- Pricing
- Freemium, Subscription, Usage-based
- Target Market
- CTOs, Enterprise Developers, AI Engineers, Product Engineering Teams
- Deployment
- API-based, Cloud-only
- Team Size
- 1-10
- Customers
- 800+ (founder's claim, October 19, 2026; 400+ reported in August 2026)
Key Features
- ✓Scrape and crawl endpoints
Return HTML, Markdown, screenshots or CSS-parsed fields, and crawl up to 500 pages with depth and URL-pattern controls.
- ✓Schema-driven extraction and Answers
Send a URL or task plus a JSON shape and get typed data back, which slots directly into agent tool calls.
- ✓Brand, company and people data
Logos, colours, fonts, styleguides, company profiles and people enrichment from a domain, the original Brand.dev capability.
- ✓Web and news search
Ranked web search with domain and freshness filters plus live and historical news search with cursor pagination.
- ✓Monitors with signed webhooks
Page, sitemap and extract monitors notify you of changes or let you pull diffs, so agents can react to updates.
- ✓Managed rendering and proxies
JavaScript rendering on every request, automatic proxy escalation and retries; failed or blocked requests are not billed.
- ✓SDKs, MCP server and CLI
Apache-2.0 SDKs for TypeScript, Python, Ruby, Go and PHP plus an MCP server, so agents can call it without glue code.
Capabilities
Use Cases
- •Agent web access
A team connects the MCP server to its coding or research agent so it can search, scrape and read live pages as Markdown.
- •Customer onboarding pre-fill
A SaaS product takes a new user's domain and fills in logo, brand colours and company details, cutting onboarding friction.
- •Structured data pipelines
A data team sends product or listing URLs with a JSON Schema and receives typed records without maintaining per-site parsers.
- •Competitive change monitoring
A product marketing team sets monitors on competitor pricing pages and sitemaps and receives webhook alerts when they change.
Ideal For
Best For
- ✓Giving coding or research agents live web access through an MCP server instead of a home-built scraper
- ✓Pre-filling onboarding with a customer's logo, colours, fonts and company profile from just a domain
- ✓Extracting structured JSON from arbitrary pages against your own JSON Schema for RAG or tool calls
- ✓Watching competitor pages, sitemaps or pricing for changes with signed webhooks
Not Ideal For
- ✗Workloads that need logged-in or authenticated pages: the founder said on Hacker News that logins are not supported yet
- ✗Teams that must self-host or keep scraping inside their own network, since there is no self-hosted edition
- ✗Organisations with strict sourcing rules on residential proxies or robots.txt compliance, which Hacker News commenters questioned at launch
Integrations
Deployment
Market & Ratings
800+ (founder's claim, October 19, 2026; 400+ reported in August 2026)
Market Analysis
Pros
- ✓Published, transparent per-request pricing with a free tier and proxies included
- ✓Broad surface in one API, from Markdown scraping to brand data and change monitors, plus SDKs in five languages and an MCP server
- ✓Hacker News users praised the docs, the responsive founder and fast time to production (one reported being live within an hour)
Cons
- ✗No support for logged-in pages yet, per the founder on Hacker News
- ✗Robots.txt is read but not strictly enforced and sites opt out via a block list, which drew pushback from site operators at launch
- ✗Residential proxy sourcing raised questions from commenters, a due-diligence item for regulated buyers
- ✗Very small team (four people at the August 2026 launch) behind a fast-growing customer base, so support and roadmap capacity is thin
Pricing
Free
$0
- ✓1,000 credits/month
- ✓2 concurrent requests
Developer
From $20/mo
- ✓5,000 credits
- ✓10 concurrent requests
Pro
From $90/mo
- ✓100,000 credits
- ✓100 concurrent requests
Growth
From $300/mo
- ✓500,000 credits
- ✓250 concurrent requests
Scale
From $500/mo
- ✓1,000,000 credits
- ✓500 concurrent requests
Scale 2.5M
From $950/mo
- ✓2,500,000 credits
- ✓500 concurrent requests
Enterprise
Contact for pricing
- ✓2.5M+ credits
- ✓Custom concurrency
- ✓99.9% uptime SLA
- ✓SSO and custom rate limits via sales
Credit-metered and published. A scrape costs 1 credit, browser actions 2, PDF OCR adds 1 per page, Answers costs 10 (fast) or 100 (Ultra), people search 10 per non-empty page and company search 2 or 10. Proxies are included, failed or blocked requests are free, paid plans allow metered overage, annual billing gives two months free, and SSO and SLAs sit behind Enterprise. One Hacker News commenter still called it expensive at launch.
Security & Compliance
Connect
Sources
This page was written from 6 sources, 5 on domains other than context.dev.
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
TestGrid ONE
AI testing agents plus real-device and browser infrastructure, deployable on-premises or air-gapped
Ampersand
Integration infrastructure that lets AI agents read and write in customer CRMs and ERPs
OutSystems Agent Experience
Governed AI development on OutSystems for Claude Code, Cursor, Codex and Kiro, via MCP
IBM Bob
IBM's agentic coding IDE and CLI for governed Java, IBM i and mainframe modernization, now self-hostable and air-gapped