Headless browsers, crawlers, and scraping APIs — the data layer for AI agents.
Firecrawl turned /scrape into a single front door for web pages, documents, and external provider data, while Apify primed its Python SDK and Crawlee for a 4.x transition.
Firecrawl introduced Fireround, an hourly-refreshed funding data provider covering 8,000+ investors and $10B+ in weekly deals, available through Alexandria in ChatGPT and Claude as well as the Firecrawl API and MCP.
Fireround is a new funding data provider built by Firecrawl that refreshes hourly and covers over 8,000 investors and $10B+ in weekly deals, with each round linking back to its sources. It is available through Alexandria in ChatGPT and Claude, as well as the Firecrawl API or MCP.
Parallel is now available as a search provider in the Browserbase Search API, with Parallel Fast for sub-second results and Parallel Advanced for deeper multi-step research. Parallel Advanced is the default, and provider and mode can be switched per request.
Firecrawl launched a Marketing Data Pack in Alexandria that wires Similarweb, Oriane, Firecrawl Trends, Particle, Icons Computer, and the Wayback Machine into a single connection for ChatGPT and Claude.
The Marketing Data Pack connects Similarweb, Oriane, Firecrawl Trends, Particle, Icons Computer and Wayback Machine through one Firecrawl connection, giving ChatGPT or Claude access to competitor traffic, acquisition channels, rising searches, social posts, creators, podcasts and visual references. It combines with Firecrawl's branding format to apply colors and typography from a site or reference site.
Stagehand 4.2 adds WebMCP tools that change with the page via page.onToolsAdded() and page.onToolsRemoved(), plus native PDF exports through page.pdf() with configurable print settings. Locator operations and snapshots now have timeouts defaulting to 20s (timeout: 0 disables), and iframe and shadow DOM interactions are more reliable.
Apify SDK 4.1.0 lets Actors pass an idempotency key to charge() to prevent duplicate charges and read input from the working directory when the store has none.
Firecrawl expanded its /scrape endpoint into Universal Scrape, letting agents read dynamic pages, PDFs, images, Office documents, and third-party people, financial, and podcast data through a single call and billing key.
Universal Scrape extends /scrape to read dynamic websites, PDFs, images, Word documents, spreadsheets and presentations by URL, and to call Alexandria provider capabilities through the same endpoint using a Firecrawl key and credits. Provider data includes people and company data from Apollo, FullEnrich and Data Legion, financial data from Fiscal.ai and Benzinga, and podcast data from Particle, with some providers requiring an org admin to accept terms before the first call.
Firecrawl Alexandria now lets AI agents in ChatGPT or Claude pull fantasy rankings, projections, player and team stats, betting odds, and news into lineup decisions, with a fantasy football demo that compares a pasted roster across scoring formats.
Firecrawl Alexandria now lets AI agents in ChatGPT or Claude research fantasy sports by combining rankings, projections, player and team stats, betting odds, and news. A fantasy football demo lets users paste a roster to compare projections across scoring formats and see suggested start/sit changes.
Firecrawl opened its Alexandria data layer to people-and-company enrichment and brought it inside ChatGPT and Codex, while Crawlee hardened its browser pool against hanging page closes and task-context leaks.
Read the digest →Agents can now search for people by title, company domain, location, role, skills, and work history, look up work emails, and enrich company firmographics, funding, headcount, technologies, job postings, and news through Apollo, FullEnrich, and Data Legion using the same Firecrawl account and credits with no second API key. Discovery is free, and each tool shows its execution price before the call.
Crawlee's patch release keeps the browser pool alive when a page.close() call never settles and stops notify() from leaking a timed-out task's context into the pool loop.
Browserbase Functions now support Secrets, allowing you to securely use credentials and API keys when connecting to external services. Store secrets i...
Alexandria is now usable inside ChatGPT and Codex through the Firecrawl plugin, giving agents live web search and scrape access to 100+ data providers and Firecrawl's specialized indexes from the same conversation. Discovery and inspection are free, while execution is billed at each tool's listed price.
Agents can now pause mid-run for human-in-the-loop steps, then resume in the same session. Set a natural-language pauseWhen condition on the run; th...
Agents in Claude and Claude Code can now reach 100+ data providers through Alexandria alongside Firecrawl's live web search, scraping, and specialized indexes, discovering a source and inspecting its inputs and price before retrieving data. Provider discovery and tool contracts are free; data calls use Firecrawl credits at the price shown on each tool.
Firecrawl launched Alexandria, a shared knowledge library that lets AI agents search curated indexes and query official data providers, and closed a $75M Series B led by Smash Capital.
Alexandria gives AI agents a common way to discover sources and retrieve information through Firecrawl, spanning the live web, official data providers, site-specific connectors and its own indexes, with agents able to search Research, Developer and Government indexes and query data providers. Firecrawl raised a $75M Series B led by Smash Capital with participation from Altos Ventures, Nexus Venture Partners, Y Combinator, Freestyle and Offline Ventures.
Webhooks are now available for Browserbase Functions. Create a Webhook in the Browserbase dashboard and subscribe to events like Function pending, run...
Search gives your agents read access to the internet, extending their context with the most up-to-date information. Search is now natively integrated ...
Firecrawl
Apify