Feed aggregator

Article URL: https://gavinray97.github.io/blog/design-by-contract-and-effects-for-llms

Comments URL: https://news.ycombinator.com/item?id=49148299

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=YJg02ivYzSs

Comments URL: https://news.ycombinator.com/item?id=49148289

Points: 1

# Comments: 0

Article URL: https://www.sciencedaily.com/releases/2026/08/260801042811.htm

Comments URL: https://news.ycombinator.com/item?id=49148264

Points: 2

# Comments: 2

Article URL: http://www.os2museum.com/wp/maximum-overkill/

Comments URL: https://news.ycombinator.com/item?id=49148258

Points: 1

# Comments: 0

Article URL: https://en.wikipedia.org/wiki/Bubble_memory

Comments URL: https://news.ycombinator.com/item?id=49148176

Points: 5

# Comments: 0

Scraping modern websites has become a massive headache. You basically have two choices: pay for an expensive API like Firecrawl/Browserbase, or run a fleet of headless Chrome instances that eat 1GB of RAM per page and still get blocked by Cloudflare.

I built Draco to fix this. It’s a fast, single-binary web scraper written in Rust. You point it at a URL, and it spits out perfectly clean Markdown or structured JSON for LLMs.

The secret sauce is that it doesn't just boot a browser for every request. It uses a tiered escalation engine:

Tier 1 (Stealth Fetch): Draco uses a custom TLS/JA4 fingerprint to perfectly mimic a real browser's network signature at the packet level. It turns out a lot of anti-bot walls will let you right through if your handshake looks correct. In my benchmarks against sites like Cloudflare and Target, Playwright ate ~500MB of RAM and timed out. Draco bypassed them in under a second using just 20MB of RAM.

Tier 2 (V8 Isolate): If it hits a React/Next.js SPA that needs rendering, Draco boots an in-process V8 engine in single-digit milliseconds. It hydrates the DOM and intercepts the hidden JSON APIs the page is calling—giving you the raw data without the overhead of a graphical browser.

Tier 3 (Real Browser): If it hits an absolute wall, it seamlessly falls back to detecting and driving a real browser on your machine.

I also built in all the tooling to make it a complete drop-in replacement for the hosted services:

Daemon Mode: Run draco serve and you get a persistent HTTP server with a Firecrawl-compatible REST API. You can swap out your API keys and self-host immediately.

Built-in MCP Server: It natively exposes a Model Context Protocol server so you can plug it directly into Claude Desktop or your AI agents.

Web Search: Built-in parallel multi-engine web search (bypassing the need for a Google Search API key).

Interact Mode: Drive a page statefully like a devtools console, persisting cookies across navigations(for LLM's mainly).

It’s completely open source (MIT/Apache-2.0). I just wanted to put this out there for anyone tired of fighting headless Chromium or paying per-page scraping costs. Grab the binary and throw a difficult URL at it.

Note that it's still a WIP so there might be some unexpected breakages of uncommon sites but for the most part its quite capable, it can handle cf-protected sites and heavy SPA's while everything else fails partially or completely while taking longer or more resources. (tested on example.com, hackernews, cloudflare, glassdoor, bluff.com, target.com, stake.com and thrill.com)

┏━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━┳━━━━━━┳━━━━━━━━━━┳━━━━━━━━━┓ ┃ Rank ┃ Tool ┃ Score ┃ Pass ┃ Avg Time ┃ Avg RAM ┃ ┡━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━╇━━━━━━╇━━━━━━━━━━╇━━━━━━━━━┩ │ #1 │ Draco │ 769.7 │ 8/8 │ 3.45 │ 216.50 │ │ #2 │ Obscura │ 384.5 │ 4/8 │ 2.68 │ 87.59 │ │ #3 │ BrowserOxide │ 373.4 │ 4/8 │ 6.42 │ 105.95 │ │ #4 │ Playwright │ 342.2 │ 4/8 │ 1.71 │ 535.07 │ │ #5 │ Bouncy │ 196.6 │ 2/8 │ 0.59 │ 19.38 │ └──────┴────────────────┴───────┴──────┴──────────┴─────────┘

Repo: https://github.com/0xchasercat/draco/

Comments URL: https://news.ycombinator.com/item?id=49148163

Points: 5

# Comments: 0

Article URL: https://seanhelvey.com/tools-and-their-tools/

Comments URL: https://news.ycombinator.com/item?id=49148159

Points: 10

# Comments: 3

Article URL: https://twitter.com/anpaure/status/2083993647268405602

Comments URL: https://news.ycombinator.com/item?id=49148129

Points: 6

# Comments: 0

Article URL: https://linuxiac.com/eu-age-verification-project-mandates-hardware-bound-attestation/

Comments URL: https://news.ycombinator.com/item?id=49148128

Points: 12

# Comments: 1

Article URL: https://www.rnz.co.nz/news/science-and-technology/868752/explainer-what-is-a-data-centre-and-why-is-everyone-freaking-out-now

Comments URL: https://news.ycombinator.com/item?id=49147313

Points: 3

# Comments: 0

Article URL: https://clickhouse.com/blog/a-quadrillion-rows-across-the-three-cloud-scaling-loghouse

Comments URL: https://news.ycombinator.com/item?id=49147311

Points: 1

# Comments: 0

Article URL: https://silentwiring.com/articles/more-agents-silent-failure/

Comments URL: https://news.ycombinator.com/item?id=49147306

Points: 1

# Comments: 0

Article URL: https://boxvxsrnquhzc9jxc2e7vr.streamlit.app

Comments URL: https://news.ycombinator.com/item?id=49147273

Points: 1

# Comments: 0

A couple months ago, I was tired of keep copy-paste across different Agent Chatbot tabs, feeling like a slave for AI sessions, so I created Kota for my own work.

Kota was designed around a concept of Knights of the Round Table.

It treats human as part of a project team, and

- Runs on top of existing CLIs with existing logins, - Gives each agent a persisted, viewable and editable identity, instead of sessions - Has a file-based long-term memory, - Uses a self-managed skillset, - Works on separated worktrees as workspace, and - Chats directly with other agents in the room.

I have been using Kota for my daily work, hobby projects, as personal life assistants and to improve Kota itself, sitting down or on the go.

Starting this week, I found it has become the only AI tool I am using now. So I am very happy to release Kota as an open source project, hoping more people can share my joy of living and working with these little friends.

In the released version, you can use some most popular CLIs with Kota, keep their configurations, and invite them as team members to sit around your project table.

These now include Codex, Claude Code, Pi, OpenCode, Gemini (Antigravity CLI), Kimi Code is already under testing and will be released soon.

Some other stuff I included in Kota to make it more comfortable to use:

- Remote with Telegram bot - Auto worktree sync and conflict resolve handoff - Skillpool, skillset, global rules, project rules, all manageable - A human-agent shared drawing board - An old school, cross-project social BBS - A lite AI-powered shell - Agents can dream about you

Some next step and points of interests:

- Agent Organization Building and Training - Agent working style cross project transfer without project data. - How to measure agent identity as meta data for task routing.

White paper: https://www.kota.place/docs/research/long-term-teammates/v1/

Comments URL: https://news.ycombinator.com/item?id=49147271

Points: 1

# Comments: 0

Article URL: https://arstechnica.com/security/2026/07/likely-illegally-claude-gained-access-to-3-networks-will-anthropic-be-held-to-account/

Comments URL: https://news.ycombinator.com/item?id=49147270

Points: 1

# Comments: 0

Article URL: https://www.overcomingbias.com/p/who-are-you-descendants

Comments URL: https://news.ycombinator.com/item?id=49147266

Points: 1

# Comments: 0

Article URL: https://gilbertramirez.com/blog/your-boss-profits-from-your-excuses-heres-the-proof/

Comments URL: https://news.ycombinator.com/item?id=49147264

Points: 1

# Comments: 0

Article URL: https://ykvm.com/2026/07/swiftui-a-story-of-mediocrity/

Comments URL: https://news.ycombinator.com/item?id=49147263

Points: 1

# Comments: 0

Article URL: https://arnoldkling.substack.com/p/is-ai-progress-stalling

Comments URL: https://news.ycombinator.com/item?id=49147254

Points: 1

# Comments: 0