Context.dev (YC S26)

We just shipped self-service SSO at Context.dev (YC S26) > no sales calls. > no implementation calls. > no waiting around for us to turn it on. Just set it up and get back to building. Available now on every Scale plan. (you're always welcome to get on a call with us though)

Back to regularly scheduled programming. PSA: New case study unlocked 🔓 OpenTag (YC S26) is building an AI coworker that lives inside Slack and Microsoft Teams and handles operational work from end to end. As a core part of their enrichment infrastructure, Open Tag needed to research people and companies in real time, roles, backgrounds, teams, profiles, and sources. Instead of building and maintaining its own crawling and parsing infrastructure, the team gave its agent our integration prompt. Two minutes later, it had integrated Context.dev (YC S26) itself. Now, a user can ask about someone directly in Slack, Context.dev (YC S26) fetches and structures the live web data, and OpenTag returns a sourced answer in the same thread. No separate research workflow. No raw crawl output. No crawling infrastructure to maintain. Just live web context, exactly when the agent needs it.

PSA: New case study unlocked 🔓 Pad is building a multiplayer AI workspace where teams collaborate with agents across shared conversations, documents, and connected tools. As they added more MCP integrations, they needed a simple way for their agents to pull real web context directly into those workflows. They connected the Context.dev (YC S26) MCP. From creating an account to their first successful retrieval: under 10 minutes. Their first use case was pulling real company logos directly into a conversation, but the larger unlock is much broader: Context.dev becomes a reusable web-context layer that Pad’s agents can call whenever a workflow needs live company data or assets. No custom retrieval pipeline. No sending users elsewhere. The agent simply gets the context it needs, exactly where the work is already happening. Full case study in the comments 👇

PSA: New case study unlocked! Instant helps more than 22,500 Shopify brands build converting pages with AI agents. But to create pages that actually feel on-brand, those agents need more than a logo. They need colors, fonts, page structure, HTML, and visual references. Instant integrated Context.dev (YC S26) to solve this. The integration was running in under an hour. Within a day, they were fully production-ready, with noticeably better output quality and developer experience. Now Instant uses our APIs to: > Generate Brand Kits that keep every page on-brand > Turn any URL into usable inspiration through HTML and screenshots > Give its agents the context they need through one API Full case study in the comments 👇

Give your agents the internet!

PSA: New case study unlocked Aldena lets you staff teams of AI agents with real roles, watch them work, and approve what ships They needed one web search + fetch layer across every model "It was super easy. I just pointed Claude at the documentation, and it figured out everything in about 30 minutes" - Aaron Delasy

How do you know you’re building something people actually want? One good signal: a customer buys a second MAXED-out plan because the rate limits on the first still weren’t enough. That’s how I met Cameron, one of the founders of Bystreet (YC W24). Bystreet helps hotel brands and vendors find the actual owners and decision-makers behind properties. They use Context.dev (YC S26) to scrape hotel websites at scale and power agents researching each hotel’s ownership structure. Cameron asked Codex to integrate us. More than 1,000,000 scrapes later, they haven’t had an issue. No scraping infrastructure to babysit. It just works. We eventually merged their accounts and worked out a better rate for them. If you need to scrape millions of pages a month, shoot me a DM. Also, they’re chill AF. 10/10 YC alums.

Introducing the Vercel Eve x Context.dev (YC S26) integration. One command gives Eve agents 34 tools for live web search, scraping, crawling, structured extraction, document parsing, monitoring, and more. No custom wrappers. No scraper fleet.