Apify blog shows how task examples can boost Actor traffic and search visibility
A blog post on Apify describes how adding task examples to an Actor doubled organic traffic and increased search impressions tenfold in a month.
Open Source & Tooling
Apify blog walks through building a live RAG pipeline with n8n and Qdrant
Apify published a tutorial showing how to crawl a website, sync changed pages into Qdrant via n8n, and re-embed only the diffs, with measured run times and cost figures.
Apify publishes tutorial on building a documentation chatbot with LangChain, OpenAI, and Pinecone
Apify released a guide showing how to combine its web scraping platform with LangChain, OpenAI, and Pinecone to create an AI-powered documentation chatbot.
Browserless compares Selenium, Playwright, and Puppeteer in new blog post
Browserless published a blog post comparing the three major browser automation frameworks.
Browserless introduces the Browser Automation Protocol
Browserless has announced a new protocol for browser automation.
Browserless launches Cursor plugin for AI-assisted browser automation
Browserless has released a plugin for the Cursor AI code editor that integrates its headless browser service directly into the development workflow.
Draco: A self-hostable, single-binary Firecrawl alternative built in Rust
A developer released Draco, an open-source, single-binary web scraping tool in Rust designed as a self-hostable alternative to paid APIs like Firecrawl and Browserbase.
Firecrawl releases Convex component for web scraping integration
Firecrawl announced a new Convex component that allows developers to integrate web scraping capabilities directly into their Convex applications.
Google Details AI-Assisted Rewrites of C/C++ Dependencies to Rust for Memory Safety
Google published a blog post on its bug hunters portal describing how it uses AI to assist in rewriting C and C++ dependencies into Rust at scale.
Indie space game SEEDS includes a built-in RISC-V emulator capable of running Linux
A small indie studio with demoscene roots has built a RISC-V emulator into their upcoming space terraforming game SEEDS, allowing players to run Linux inside the game.
Jactl brings virtual threads to Java 8 without Project Loom
The Jactl scripting language introduces virtual thread support for Java 8, bypassing the need for Project Loom.
LuaCAD brings parametric CAD scripting to Lua with operator overloading
A new tool called LuaCAD lets users model solids in Lua instead of the OpenSCAD language, offering a CLI and desktop app with preview and editor.
Lunatik 5.0 adds eBPF integration for scripting the Linux kernel with Lua
Lunatik 5.0, a framework for scripting the Linux kernel with Lua, now integrates with eBPF.
Magic.dev publishes research on compute-efficient pretraining for trillion-parameter models
Magic.dev has shared a blog post detailing techniques for compute-efficient pretraining and scaling to trillion-parameter models.
Mini-AGI: A dynamic continual learning model trained on 8GB VRAM appears on GitHub
A developer released Mini-AGI, a dynamic continual learning model that runs on 8GB VRAM, on GitHub as a Show HN post.
PageSieve browser extension launches for Firefox-based web scraping
A new open-source browser extension called PageSieve enables users to scrape data from websites directly within Firefox.
Playwright 1.63.0 adds test locks for safe concurrent access to shared resources
Playwright 1.63.0 introduces named test locks that prevent concurrent execution of tests sharing the same lock name across files, workers, and projects.
Playwright v1.64.1 patches Electron window loss and Chromium cross-origin redirect regression
Microsoft released Playwright v1.64.1, a patch fixing two bugs: an intermittent loss of a packaged Electron app's first window and a Chromium 155+ regression that aborts preflighted cross-origin redirects when any route is registered.
Puppeteer 25.10.0 adds video-stream screen recording and Firefox 155.0 support
Puppeteer released version 25.10.0 of puppeteer-core, introducing a video-stream-based screen recording feature via page.record() and rolling to Firefox 155.0.
Puppeteer browsers v3.2.5 patches symlink escape and path traversal vulnerabilities
Puppeteer released v3.2.5 of its browsers package, fixing a symlink chain escape during zip extraction and a path traversal issue from buildId, along with a fix for alias removal on uninstall.
Puppeteer v25.13.0 adds ConsoleMessage methods, WebSocket payload config, and fetch request/response conversion
Puppeteer released version 25.13.0 of puppeteer-core with new features including page, frame, and worker methods on ConsoleMessage, configurable WebSocket max payload, and asFetchRequest/asFetchResponse methods on HTTPRequest and HTTPResponse.
Puppeteer v25.14.0 ships network emulation migration and Firefox 158.0 support
Puppeteer released version 25.14.0 of puppeteer-core, migrating network emulation to a new rule-based API and rolling to Firefox 158.0.
Puppeteer v25.15.0 adds handshake timeout, tab info, and geolocation override removal
Puppeteer released version 25.15.0 of puppeteer-core with several new features and fixes.
Puppeteer v25.16.0 adds unmodified post data access and Chrome 156 roll
Puppeteer released version 25.16.0 of puppeteer-core with support for unmodified post data via HTTPRequest.content(), base64 decoding, a roll to Chrome 156.0.8078.47, and a fix for page ranges without spaces in Page.pdf().
Puppeteer v25.8.0 adds followSymlinks option and Windows detached mode fix
Puppeteer released version 25.8.0 of puppeteer-core with a new followSymlinks option, a validatePath fix for computeSystemExecutablePath, and a change to launch browsers in detached mode on Windows to reduce flakiness.
Puppeteer v25.9.0 rolls Firefox 154.0 and Chrome 152.0, adds shadow root support in MutationPoller
Puppeteer released version 25.9.0 of puppeteer-core, updating browser versions and adding support for shadow roots in MutationPoller and tool execution cancellation in webmcp.
PyScrappy brings self-healing selectors and an MCP server to web scraping
A new open-source Python library, PyScrappy, offers self-healing CSS/XPath selectors and an MCP server for resilient web scraping.
RubyLLM 2.x adds step-by-step agent loops, human-in-the-loop tool approval, and persistent agents
SerpApi's blog covers the release of RubyLLM 2.0 and 2.1, which introduce incremental agent loop execution, human approval for tools, and database-backed agent persistence.
Scrapfly compares 7 C# web scraping libraries across parsing, automation, and orchestration layers
Scrapfly published a workflow-based comparison of seven C# web scraping libraries, covering HTML parsing with Html Agility Pack and AngleSharp, browser automation with Playwright for .NET, Selenium, and PuppeteerSharp, and crawl orchestration with Abot and DotnetSpider.
Scrapfly compares 8 Python HTTP clients for web scraping in 2026
Scrapfly published a blog post evaluating eight Python HTTP clients on async support, HTTP/2, HTTP/3, TLS impersonation, and maintenance, with runnable examples.
Scrapfly compares 9 Go web scraping libraries and APIs for 2026
Scrapfly published a comparison of nine Go web scraping libraries and APIs, evaluating each by its best use case, current release, and license.
Scrapfly compares Browser Use and Playwright for web scraping
Scrapfly published a blog post comparing Browser Use and Playwright, covering architectural differences, speed and cost tradeoffs, silent failure risks, and a hybrid approach for production scraping.
Scrapfly compares five open-source Amazon scrapers on GitHub
Scrapfly published a blog post evaluating five open-source Amazon scraper repositories by stars, commit recency, license, code paths, and live failure modes.
Scrapfly compares six modern command-line tools as alternatives to cURL and Wget
A blog post from Scrapfly evaluates HTTPie, aria2, and other tools that address specific limitations of cURL and Wget, including a managed fetch tool for blocked requests.
Scrapfly publishes a lookup catalog for Scrapy error messages
Scrapfly released a blog post cataloging Scrapy error messages with diagnostic checks and version-specific fixes.
Scrapfly publishes guide on testing web scrapers before production
Scrapfly released a blog post detailing methods for building parser tests, HTML fixtures, live checks, and deployment gates to catch silent scraper failures early.
Scrapfly publishes job-by-job guide to Puppeteer alternatives for web scraping
Scrapfly released a blog post comparing seven alternatives to Puppeteer for web scraping, covering library rewrites, crawl frameworks, no-browser approaches, and remote stealth browsers.
Scrapfly ranks five open-source LinkedIn scrapers on GitHub by auth model and ban risk
Scrapfly published a blog post evaluating five open-source LinkedIn scraping repositories on GitHub, ranking them by authentication approach, maintenance status, and real-world blocking risk as of August 2026.
Scrapfly ranks four open-source proxy scrapers still viable in 2026
A blog post from Scrapfly filters the crowded open-source proxy tool landscape down to four actively maintained scrapers and checkers worth using this year.
Scrapfly ranks six open-source YouTube scrapers by job, flags two failures
Scrapfly published a blog post comparing six open-source YouTube scrapers, including a GitHub snapshot from August 11, 2026, and noting two projects that failed in their tests.
Scrapfly ranks top JavaScript web scraping libraries for 2026
Scrapfly published a guide ranking eight JavaScript and Node.js libraries for web scraping, covering HTTP clients, parsers, headless browsers, and the Crawlee framework.
Scrapfly reviews nine Scrapy extensions and middlewares for 2026
Scrapfly tested nine Scrapy extensions and middlewares against Scrapy 2.18, covering rendering, TLS fingerprints, proxies, extraction, shared queues, and deployment.
Scrapfly rounds up top open-source Facebook Marketplace scrapers on GitHub for 2026
Scrapfly published a blog post listing the five best open-source Facebook Marketplace scrapers on GitHub as of 2026, along with repos to avoid.
Scrapfly surveys six Rust web scraping libraries and APIs for 2026
A blog post from Scrapfly rounds up six Rust-based tools for managed page access, static HTML parsing, article extraction, and browser control.
ScrapingBee covers Ruflo, a multi-agent orchestration layer for Claude Code and Codex
A blog post on ScrapingBee introduces Ruflo, an open-source meta-harness that coordinates swarms of AI agents by wrapping around Claude Code or Codex.
ScrapingBee publishes guide on using CloudProxy for web scraping
ScrapingBee's blog post covers setting up CloudProxy, its limits, and its managed API for lightweight scraping tasks.
ScrapingBee publishes tutorial on sharding Playwright tests with Testkube on Kubernetes
A new tutorial from ScrapingBee walks through distributing Playwright browser tests across Kubernetes pods using the --shard flag and a Testkube agent.
ScrapingBee publishes tutorial on using Playwright in Ruby
ScrapingBee released a guide covering installation and usage of the community-maintained playwright-ruby-client gem for web scraping and testing in Ruby.
Scrapy 2.18.0 ships with httpx2-based download handler and stable HTTP/2 support
Scrapy 2.18.0 replaces the Twisted-based HTTP/2 handler with httpx2, makes brotli and Zstandard compression always available, and changes late crawler attribute access to raise RuntimeError.
Scrapy 2.19.0 ships RemoteControl extension and experimental aiohttp download handler
Scrapy released version 2.19.0, introducing a RemoteControl extension for HTTP-based crawl inspection and control, alongside an experimental aiohttp download handler now set as the default for reactor-less runs.
SerpApi launches Review Audit, an open-source tool to assess Google Maps rating authenticity
SerpApi released Review Audit, an open-source tool that analyzes a Google Maps place's recent reviews and reviewer histories to estimate how much of its rating is attributable to its own reviewers.
Spotatui revives unmaintained Spotify TUI with workarounds for web API rate limits
A developer has revived the abandoned spotify-tui project as Spotatui, a terminal-based Spotify client that bypasses web API rate limits.
Stagehand decouples Browse releases from SDK release pipeline
Stagehand separates Browse release management from the Stagehand SDK release group, introducing an independent publishing job for Browse.
Stagehand exposes browser-side function waits across SDKs
Stagehand released a pre-release version that exposes browser-side predicate waiting consistently across its SDKs, eliminating transport polling.
Stagehand exposes PDF rendering across SDKs
Stagehand released version 4.2.0a0.dev1543 of its Python SDK, adding PDF rendering support through existing SDK conventions.
Stagehand Python 4.0.3a0.dev1483 adds stagehand_facade tool surface for eval benchmarking
A dev release of Stagehand Python introduces a facade tool surface that allows evals to benchmark the exact byte-identical interface shipped to Claude Code, Codex, and Pi integrations.
Stagehand Python 4.2.0a0.dev1528 restores V3 network capture via CDP sidecar
Stagehand released a dev build that reimplements Browse V3 network capture for the Python SDK using a CLI-private CDP sidecar, avoiding changes to the core protocol or public SDKs.
Stagehand Python 4.2.0a0.dev1533 adds CLI secrets management for Browse Cloud
Stagehand released a dev build that introduces CLI commands to manage project and function secrets for Browse Cloud.
Stagehand Python 4.2.0a0.dev1591 fixes waitForSelector timeout tracking
A development release of Stagehand Python fixes waitForSelector to properly account for frame resolution and locator readiness within its timeout.
Stagehand Python adds event hooks for WebMCP tool lifecycle
Stagehand Python 4.2.0a0.dev1524 introduces event hooks that fire when WebMCP tools are added or removed from a page.
Stagehand Python adds timeout parameter to all locator methods
Stagehand Python version 4.2.0a0.dev1556 exposes a configurable timeout parameter across all 17 terminal locator methods, with a default of 20 seconds.
Stagehand Python v4 alpha restores server-side caching as default
Stagehand Python v4.2.0a0.dev1530 flips server-side caching from opt-in back to on-by-default, matching v3 behavior after an upgrade gap silently dropped caching for users who did not opt in.
Stagehand Python wires progress tracker into act, extract, and observe methods
Stagehand Python v4.2.0a0.dev1576 refactors act(), extract(), and observe() to use a unified progress tracker instead of per-method timeout tracking.
Stagehand releases Browse 0.11.0 with secret attachment support
Stagehand published version 0.11.0 of its Browse CLI, decoupling it from the Stagehand SDKs and adding a function to attach existing project secrets by ID.
Stagehand releases browse@0.11.1 with project scaffolding via functions init
Stagehand published browse@0.11.1, decoupling the CLI from the SDK and adding a functions init command that scaffolds a Stagehand project using @browserbasehq/stagehand instead of playwright-core.
Stagehand Releases Browse CLI 0.10.0 with V4 Runtime Migration
Stagehand published version 0.10.0 of its Browse CLI, migrating the runtime to Stagehand V4 and removing the --return-xpath option from coordinate actions.
Stagehand releases Browse CLI v0.10.0 with V4 runtime migration
Stagehand published Browse CLI version 0.10.0, decoupling it from the Stagehand SDKs and migrating its runtime to Stagehand V4.
Tilion launches Fortress v3, a stealth Chromium engine for undetected web agents
Tilion (YC F26) released version 3 of Fortress, its stealth Chromium-based browser infrastructure, as a free open-source tool for developers.
Vessel 0.3 brings pluggable drivers and middleware pipeline to Ruby crawling
SerpApi covers the release of Vessel 0.3, a Ruby crawling framework that adds pluggable drivers, a fields API, middleware pipeline, proxy rotation, cookies, retries, and callbacks.
Zyte compares three browser setups for Scrapy: Playwright, Patchright, and Zyte CDP
Zyte published a blog post demonstrating how a single Scrapy spider can run unchanged across three browser backends: stock Playwright, Patchright via a new hook, and Zyte's remote CDP browser.
Zyte details Scrapy 2.19's RemoteControl extension for live crawl inspection
Zyte published a technical walkthrough of Scrapy 2.19's RemoteControl extension, which runs an authenticated HTTP server inside each crawl for live inspection and control.
Zyte launches 'Modern Scrapy for experienced developers' tutorial series
Zyte published the first part of a new blog series aimed at experienced developers building production-ready Scrapy projects.
Zyte publishes uv 0.12 cheatsheet covering changed defaults and command-line traps
Zyte released a practical cheatsheet documenting the default changes in uv 0.12, including project structure, hash checking, and lockfile behavior.
Zyte releases scrapy-spidey-sense, a preflight CLI for Scrapy projects
Zyte has open-sourced a command-line tool that performs static analysis on Scrapy projects before a crawl begins, scoring production-readiness and linking findings to fixes.