Users expressed confusion about whether the product is related to Bun or Node.js based on the title and instructions. Additionally, some users felt that the library might be unnecessary if it only consists of 92 lines of code.

Users criticized the product for having a mismatch between the title and instructions, and for including an unnecessary library.

Questions about project features and Cloudflare bypass.

What makes this project more undetectable?

Nimble API's Product Hunt launch garnered positive feedback, with users praising its free trial, API access, and ease of use for data scraping. Several users expressed interest in using it for web crawling and automating CRM workflows. Questions were raised about legal and ethical considerations, compliance, and responsible data collection. Users also inquired about pricing compared to alternatives like Jina and the product's development timeline. Many congratulated the team on the launch and expressed eagerness to try the API.

Users expressed concerns regarding the product's data collection practices and requested more transparency. The pricing was also criticized, with some users finding it expensive compared to alternatives like Jina Reader. These were the primary areas of concern raised during the Product Hunt launch.

FlowScraper's Product Hunt launch garnered positive feedback, with users calling it a game-changer, especially for non-technical users. Key discussion points included inquiries about API availability, accessing generated code, and anti-bot measures on complex sites like React websites. Users requested e-commerce and LinkedIn templates and guides. There were also questions about Datadome compatibility, lifetime access, self-deployment, and the sophistication of anti-bot protection beyond existing solutions.

Users criticize the use of Unreal Engine blueprints for web scraping. They also suggest improvements, such as adding e-commerce templates, a LinkedIn page, and offering a free year. Furthermore, some users reported that the anti-bot measures are insufficient, with some sites failing when using puppeteer-extra-plugin-stealth.

The Show HN submission for AgentQL, an AI-powered semantic framework for web interaction, has received minimal engagement, with no substantive content available in the comments. The majority of comments have been flagged and are pending review, indicating either spam, inappropriate content, or other violations of community guidelines.

Users appreciate the tool, highlighting its usefulness in transforming webpages into structured data and suggesting improvements like handling JavaScript-heavy sites and adding Markdown output. Concerns about cost and efficiency are noted, with suggestions for reusable scripts and compatibility with OpenAI's API. Some users report specific use cases like CSS selector generation and structured data extraction with LLMs. Challenges such as antibot measures and captchas are mentioned, as well as a curiosity about the tool's prompt and capabilities like instruction following and screenshot parsing.

Users criticized the Show HN product for lacking a reusable script for LLM, high costs associated with calling LLM each time and scaling content size, and concerns about cost and hallucination frequency. Other issues include difficulty with information hidden in text, the need for handling JavaScript sites, lack of Markdown output, and latency issues with web LLMs. Users also suggested solving underlying problems such as antibots and captchas.

Users appreciate the Show HN product for its ability to package assets into a single binary, similar to RedBean and Pocketbase, and its cross-platform functionality. There's interest in single-file webpages, with comparisons to .mht files and SingleFile extension. Some discuss the effectiveness on static sites and issues with subsites. The conversation includes technical suggestions like status codes and command-line options, and there's a request for license addition, with MIT and BSD mentioned. Users also discuss the tool's utility for website impersonation, backup, and static hosting, with some confusion about deployment and SSR. A spam link and a dead comment are noted.

Users criticized the product for its lack of support for multiple browsers (Firefox, Safari) and formats (MHT, PDF, .riv files), unclear functionality and benefits, and potential copyright issues. There were also concerns about privacy risks with HAR files, inefficiency with large files, and incorrect error handling. The product's slow performance, error messages, and lack of clear documentation on saving/restoring tasks and deployment were also noted. Some users were concerned about the tool being useful to scammers, while others questioned the necessity of the tool for copying websites.

Question about using Redis over Postgres.

High maintenance surface area.

Users discuss Crawlee's usability, comparing it to Scrapy, Selenium, and other scraping tools. They praise its easy setup, performance, and configurability, but note documentation issues and a lack of examples. Some inquire about specific features like web scraping opt-out, 2FA handling, and anti-blocking. Others question the ethics of scraping and the respect for host settings. There's interest in Python support and requests for code snippets in documentation. The tool is noted to be free, open-source, and Python-compatible, with some users planning to try it for projects.

Users criticized the product for inadequate documentation, particularly missing test case snippets and unclear feature descriptions. The coding style and lack of unique features were also noted. Concerns about ethical considerations, such as web scraping opt-out protocols and bot detection avoidance, were raised. Technical issues mentioned include insufficient type hint coverage, customization difficulties, and inadequate handling of proxies, caching, and 2FA. Comparisons with Scrapy and Crawlee suggest confusion about the product's unique value proposition, and there were requests for better automation and examples, especially for dynamic sites and non-HTML content.

Links to previous discussions on Hacker News.

Users have expressed concerns about Flyscrape's effectiveness in scraping real-world sites, particularly those with anti-scraping measures like Cloudflare. There's a demand for more features, dynamic settings, and examples of advanced web-scraping. Comparisons with other tools like Colly and Playwright are requested, and there's interest in handling dynamic content and site changes. Some users find the tool potentially useful and plan to test it, while others have moved to different technologies. There are also technical questions about specific functions and suggestions for improvements, including opening the core for feedback.

Users criticized the product for its inability to handle real-world scraping challenges such as dynamic content, JavaScript rendering, and anti-scraping measures. The lack of examples, broken documentation links, and the need for clearer methods were also mentioned. Some noted the inefficiency of Python compared to UNIX utilities and the necessity to spoof user agents. The product's inability to handle iframes, shadow elements, and the absence of programmatic extensibility for custom browser control were highlighted. Users also pointed out the risk posed by gatekeeping technologies like Cloudflare and Akamai.

Users are interested in the product, with one asking about async handling similar to waitFor in puppeteer. Another user compares it to colly, expressing satisfaction but openness to improvements. A third user appreciates the neat approach but initially missed that it's written in JavaScript.

The main criticism is the lack of information on asynchronous handling in the readme. Additionally, there is a general openness to improvements.