releases.shpreview

Browser Automation & Scraping

Headless browsers, crawlers, and scraping APIs — the data layer for AI agents.

This week: Firecrawl edges out Browserbase's WebMCP in a week of agent-first tooling →
ThuJul 16, 20261 release
Browserbase
Daily summary

Stagehand silences AI SDK warnings and improves browser skill sourcing

Stagehand fixed a noisy AI SDK warnings in act/extract/observe and corrected how browser skills are sourced from the CLI package.

  • Stagehand silenced the "system message in messages" AI SDK warning that appeared during act, extract, and observe operations
  • Stagehand fixed the browser skill source path to correctly resolve from the packages/cli browse SKILL.md file
BrowserbaseBrowserbase1
Stagehandstagehand-server-v3/v3.7.3
MonJul 13, 20261 release
Browserbase
Daily summary

Stagehand adds domain access controls

Stagehand v3.7.0 lets users restrict browser navigation to an allowlist of domains and introduces a simple or condition syntax for element matching.

  • Stagehand’s new setDomainPolicy can lock browsing to specific domains via allowedDomains, giving teams a way to enforce scraping boundaries
  • Element queries now support or conditions in CSS and XPath selectors, reducing the need for fallback logic when multiple element signatures are possible
BrowserbaseBrowserbase1
Stagehand@browserbasehq/stagehand@3.7.0
TueJun 30, 20261 release
Firecrawl
Daily summary

Firecrawl launches continuous web surveillance

Firecrawl's new /monitor endpoint searches the web on a schedule and alerts via webhook or email when new results match a user-defined goal.

  • Firecrawl's /monitor watches the full web continuously, alerting when a search query finds something that meets a plain-English goal
  • Separate queries and goal fields let users tune recall (what to search) and precision (what triggers an alert), with webhook or email delivery
FirecrawlFirecrawl1
MonJun 29, 20261 release
Browserbase
BrowserbaseBrowserbase1

Introducing Browserbase Agents

Browserbase Agents is a managed solution that enables you to deploy agents on the web with a simple API call. No code to maintain or deploy—describe t...

Read post
Introducing Browserbase Agents
ThuJun 25, 20261 release
Browserbase
Daily summary

Stagehand browse screenshot saves files by default

Stagehand changed its browse screenshot command to write a file to disk by default instead of printing base64 to stdout, with a --base64 flag preserving the old behavior.

  • Stagehand browse screenshot now saves a file to the current directory by default, printing the path as JSON, and adds a --base64 flag for the legacy stdout behavior
BrowserbaseBrowserbase1
Stagehandbrowse@0.9.0
TueJun 23, 20261 release
Firecrawl
Daily summary

Firecrawl adds research index and removes API key barrier

Firecrawl 2.11.0 introduced a Research Index spanning 3M+ arXiv papers and GitHub code, made core scrape and search endpoints usable without an API key from official clients, and added automatic PII redaction.

  • Firecrawl Research Index lets you search across 3M+ arXiv papers, GitHub issues, PRs, and READMEs, outperforming alternatives by 18% on arXivQA recall
  • Scrape, search, interact, and parse endpoints now work without an API key when accessed through official MCP, CLI, or SDK clients
  • A new redactPII option strips personal data from scraped content, joined by deterministicJson output formatting and automatic video discovery on any page
FirecrawlFirecrawl1

Research Index, keyless access, and PII redaction ship

v2.11.0

Firecrawl Research Index searches 3M+ arXiv papers and GitHub code, core endpoints can be accessed without an API key from official clients, and a new redactPII option strips personal data from scraped content. Also ships deterministicJson format and video discovery on any page.

Read post
v2.11 Release
FriJun 19, 20261 release
Firecrawl
Daily summary

Firecrawl launches a dedicated research index for AI papers

Firecrawl v2.11.0 adds a specialized search index covering 3M+ arXiv papers with linked GitHub code, refreshed daily, beating the next best provider by 18% on arXivQA recall.

  • Firecrawl introduced a Research Index purpose-built for agentic AI/ML work, letting you search 3M+ arXiv papers alongside their source code repos, fetch paper details and related work, and verify claims against full text
  • Core endpoints for scrape, search, interact, and parse now work without an API key when accessed through official MCP and CLI integrations, lowering the barrier for quick automation
FirecrawlFirecrawl1
GitHub Releasesv2.11.0
ThuJun 18, 20262 releases
Browserbase
BrowserbaseBrowserbase2
Stagehand@browserbasehq/stagehand@3.6.0
MonJun 15, 20261 release
Firecrawl
FirecrawlFirecrawl1
ThuJun 11, 20261 release
Browserbase
BrowserbaseBrowserbase1
Stagehand@browserbasehq/stagehand@2.5.9
TueJun 9, 20261 release
Browserbase
BrowserbaseBrowserbase1
Stagehandstagehand-server-v3/v3.7.2
FriJun 5, 20261 release
Browserbase
BrowserbaseBrowserbase1
Stagehandbrowse@0.8.3
ThuJun 4, 20261 release
Apify
ApifyApify1
Crawleev3.17.0
WedJun 3, 20262 releases
Browserbase
BrowserbaseBrowserbase2
TueJun 2, 20261 release
Browserbase
BrowserbaseBrowserbase1
FriMay 29, 20261 release
Apify
ApifyApify1
Apify SDKv3.4.1
ThuMay 28, 20261 release
Browserbase
BrowserbaseBrowserbase1
Stagehandstagehand-server-v3/v3.7.1
WedMay 27, 20261 release
Browserbase
BrowserbaseBrowserbase1
Stagehandstagehand-server-v3/v3.7.0