Show HN за 13 декабря 2025 г.

  I've been using Claude Code daily for about 6 months. After building the same prompts over and over, I started documenting them as "recipes" - structured prompts with context about when to use them and what output to expect.

  This repo has 100 recipes covering common knowledge work tasks:

  - Meeting notes → action items
  - Status reports
  - Performance reviews
  - Proposals and presentations
  - Data analysis narratives
  - SOPs and documentation

  Each recipe includes:
  - The problem it solves
  - When to use it (and when not to)
  - Prerequisites
  - Step-by-step prompts
  - Example output
  - Troubleshooting tips

  The recipes are organized into 10 tiers from universal tasks (everyone needs meeting notes help) to specialized functions (M&A due diligence, legal research). 

  I also included 10 sample slash commands in the /premium folder that you can install directly into Claude Code's ~/.claude/commands/ directory.

  Happy to answer questions about how these evolved or discuss the patterns I've noticed in what makes prompts work well for different task types.

Long time lurker, thought I would contribute back something to the community (at least the meat eaters in the UK). One thing that's been a pain for me to get a good understanding of is what the best price is for online butchers based around the UK. I like high quality meat (including some of the bigger cuts for kamado/bbq), and I'm willing to pay a premium however there isn't a single resource that would let me get an overview of what's available and price movements over time.

For my Xmas meat I didn't want to start building out a spreadsheet so I built a price tracker with quite a few convenience features to make it easy to search for certain cuts of meat across commonly mentioned butchers on Reddit and other communities.

If you find it useful let me know if there are any features that could help to make it even easier to find the cuts of meat that you're looking for!

Here's the tracker: https://offer-spider.onrender.com

Hosted on Render, built with Nextjs and SQLite. Spidering various custom e-com sites (WooCommerce, Shopify) on a daily basis.

My current TODOs: - Search is currently free-text OR. Add AND to allow focussing down a search, and select from facets/categories/taxonomy - Common search terms to pre-fill - Exposing price over time metrics - Further consolidation of SKUs to an internal taxonomy - More butchers - Expanding to other use cases outside of meat!

Like many builders, we found ourselves constantly switching tools just to test different models or creative directions. Each platform felt limited to a single workflow or use case. So we decided to build a space that’s more modular, exploratory, and creator-first.

Kinkora focuses on:

Supporting popular generative models

Making experimentation fast and enjoyable

Laying the foundation for a creative community, not just a generator

Our long-term goal isn’t just “generate content”, but to create a place where creators can play, iterate, and discover new ideas as models and techniques evolve.

We’re early, actively iterating, and would love feedback from fellow indie builders on:

Feature direction

Community mechanics

Creator-friendly workflows

Happy to answer any questions

So I built my own with a sleek dark theme, a search that accepts any input, and zero ads or other distractions.

I’d like to share an open-source project we’ve been working on for a while: Browser4.

The motivation came from a recurring frustration: most browser automation tools (Playwright, Selenium, Puppeteer) are excellent for human-written scripts, but start to show friction when used as a core execution layer for AI agents or at very high concurrency.

So instead of building “another wrapper around Playwright”, we experimented with a different direction: designing a browser engine where AI agents are first-class citizens.

### What Browser4 is

Browser4 is a browser automation engine built on native Chrome DevTools Protocol (CDP), with a focus on:

* Coroutine-safe concurrency (designed to run many browser sessions in parallel)

* Agent-oriented APIs (navigation, interaction, extraction as composable actions)

* Hybrid extraction: ML agent driven extraction + LLM extraction + structured selectors + an SQL-like DOM query language (X-SQL)

* Low-level control without Playwright-style abstraction overhead

It’s written in Kotlin/JVM, mainly because we needed predictable concurrency behavior and long-running stability under load.

The project is fully open-source (Apache 2.0).

### What it’s not

* It’s not a drop-in Playwright replacement.

* It’s not a no-code RPA tool.

* It’s not “LLM magic” — LLMs sit outside the browser engine.

Browser4 intentionally stays close to the browser execution layer and leaves planning/reasoning to external agent loops.

### Current use cases we’re testing

* Large-scale web data extraction

* Agentic workflows (search → navigate → extract → summarize)

* Price / content monitoring with frequent revisits

* High-concurrency crawling where browser startup and context switching are bottlenecks

On a single machine, we can sustain very high daily page visits, though we’re still validating benchmarks across different workloads.

### Open questions (where I’d love feedback)

* For agentic systems, does it make sense to bypass Playwright entirely and work closer to CDP?

* Where do you see the biggest pain points when combining LLMs with browser automation today?

* Is JVM a reasonable choice here, or is Python still the better tradeoff despite concurrency limits

* What abstractions would you want in a browser engine built for AI agents?

### Links

* GitHub: https://github.com/platonai/browser4

* Website (light overview): https://browser4.io

Happy to answer technical questions or hear criticism — especially from people running browser automation or agent systems in production.

Thanks for reading.

A) Free without adverts B) Worked well on mobile/tablets C) Was clutter free and easy to use

I wondered how easily I could vibe code a solution.

Here's the result.

Everything here was "vibe coded" to an extent, the graphics, sounds, art-work, even the github pipelines ( and I used chatgpt to instruct me how to configure everything in AWS )

I'm particularly pleased with the phoneme game, https://free-kids-typing-games.com/games/phoneme-sound-lab/ which was crafted using Google's TTS engine, and I think could be expanded further.

The mobile keyboard is also particularly impressive and works really well for little fingers.

I thought the hacker news crowd might be interested.

You can see the code in all it's gory AI generated detail here

https://github.com/Alan01252/free-kids-typing-games.com

[1]: https://heroicgameslauncher.com/ [2]: https://github.com/Winetricks/winetricks

It uses a sharded architecture (one thread per symbol group) to avoid mutex contention during matching. Recently moved to int64_t fixed-point pricing and added memory compaction to prevent leaks in long-running processes.

Current tools (Google Docs, Notion, etc.) were designed for human-to-human collaboration. When I copy-paste Claude's suggestions into a doc, all attribution is lost. My team can't tell which parts I wrote vs AI-generated. In open source, this creates trust issues.

*Tandem's approach:* - Every edit is tagged as Human or AI - Git-based version control (full history, not just "last modified") - Real-time collaboration using Yjs CRDT - MCP integration (AI can directly edit documents using tools)

*Tech stack:* React 19, TipTap, Yjs, Hono, Bun, MCP

*Live demo:* https://tandem.irisgo.xyz

Think "Google Docs meets Git" but designed for human-AI teams. I'm treating this as a bottom-up approach to building an AI-native workspace (inspired by Sam Altman's recent comments about needing an "AI-native Slack").

Currently in MVP stage. Looking for feedback from the HN community: - Does this solve a problem you have? - What features would make this indispensable? - Concerns about AI attribution?

Happy to answer technical questions about the implementation!

I built an AI music generation platform that solves a problem I had as a content creator: finding royalty-free music that's actually usable.

Key features: • Generate songs up to 8 minutes (vs typical 2-4 min limits) • Multiple AI models: lyrics generation, vocal removal, stem separation • Commercial license included - safe for YouTube, Spotify, TikTok • Export in WAV/MP3 with studio quality

Technical stack: Next.js 15, multiple AI providers (DeepSeek, OpenAI, etc.), PostgreSQL, Stripe for licensing.

Free tier available with 2 songs/month. Would love feedback from the community.

Link: https://aisongcreator.app

Happy to answer any questions about the AI models, licensing approach, or technical architecture.