frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Who Is Responsible for Tackling Professional Deskilling?

https://www.degruyterbrill.com/document/doi/10.1515/humaff-2026-0080/html
1•geox•30s ago•0 comments

Anduril, Qualcomm launch BlueChip, a new kind of US-made laptop

https://www.theverge.com/gadgets/1019274/anduril-qualcomm-bluechip-laptops-american-made-military
1•kikkia•33s ago•0 comments

US agency is again probing if Polymarket broke 2022 settlement, sources says

https://www.reuters.com/business/finance/us-agency-is-again-probing-if-polymarket-broke-2022-sett...
1•tartoran•1m ago•0 comments

I turned a looped transformer into a decision model that scores answer options [pdf]

https://wyat.io/wyat2-brief.pdf
1•teilom•1m ago•1 comments

Open-source n8n and MCP workflows for UCP store audits and B2B signals

https://github.com/edtool79-collab/hirepulse-mcp/blob/main/templates/SUPER_TOOLS_PLAYBOOK.md
1•ystudio3d•3m ago•0 comments

The Diffusion of Reality: How the Interface Imperceptibly Became a Habitat

https://medium.com/@smoking2man/the-diffusion-of-reality-how-the-interface-imperceptibly-became-a...
1•Aldus551•4m ago•0 comments

Mossad Has Devices Like the Exploding Pagers in "Every Country You Can Imagine"

https://www.theyeshivaworld.com/news/israel-news/2462875/former-mossad-chief-yossi-cohen-planted-...
5•newspaper1•5m ago•0 comments

The American Divide on Empathy

https://kottke.org/26/10/the-american-divide-on-empathy
2•lehi•5m ago•1 comments

Dismissing Claude's request for feedback is itself considered feedback (2025)

https://keydiscussions.com/2025/09/29/dont-even-dismiss-the-how-is-claude-doing-this-session-prom...
1•kaycebasques•6m ago•0 comments

ZacOS 9 and Mavernix: Two classic Apple desktops find their Wayland to Linux

https://www.theregister.com/os-platforms/2026/10/27/zacos-9-and-mavernix-two-classic-apple-deskto...
1•losgehts•6m ago•0 comments

Microsoft PowerToys Meets NLP

https://grabthat.ai/blog/microsoft-powertoys-natural-language-integration
1•kevanparker•6m ago•0 comments

OSRS Wiki and RuneLite increasingly under strain from low-effort AI development

https://oldschool.runescape.wiki/w/User:Cook_Me_Plox/OSRS_Wiki_and_RuneLite_are_increasingly_unde...
1•Planktonne•7m ago•0 comments

Blanets and Weird Worlds

https://www.antipope.org/charlie/blog-static/2026/10/bizarre-blanets-and-other-weir.html
1•pavel_lishin•9m ago•0 comments

Debcraft at v1.0

https://optimizedbyotto.com/post/debcraft-1-0-released/
1•ottoke•9m ago•1 comments

Echo-3 and the SpAItial App

https://spaitial.ai/blog/echo-3-release
1•frozenseven•10m ago•0 comments

Doctor removes prostate from 35 miles away in first telerobotic surgery in US

https://www.theguardian.com/us-news/2026/oct/27/first-telerobotic-surgery
1•bookofjoe•10m ago•0 comments

I benchmarked 4 web search APIs for AI agents on 30 live queries

https://github.com/psinglap/websearch-api-eval
1•pallavisingla•11m ago•0 comments

A critical look at self-distillation for reasoning

https://jasonkena.github.io/blog/posts/does_self_distillation_improve_reasoning/
2•RevoGen•16m ago•1 comments

The Slopocalypse Survival Guide

https://alicegg.tech/2026/10/27/slopocalypse
4•zer0tonin•16m ago•0 comments

Cloud Rats and Software Roaches

https://idlewords.com/2026/10/cloud_rats_and_software_roaches.htm
2•pavel_lishin•17m ago•0 comments

Intel: Panther Lake

https://en.wikipedia.org/wiki/Panther_Lake_(microprocessor)
2•tosh•17m ago•0 comments

Claude vs. DeepSeek vs. GLM: what coding agents pick

https://whatagentspick.com/claude-vs-deepseek-vs-glm/
2•nsiradze•17m ago•0 comments

Satya Nadella Rethinks Whether More Powerful AI Is Safer

https://www.businessinsider.com/satya-nadella-ai-alignment-changed-mind-different-problem-scales-...
3•hirodxb•18m ago•1 comments

Wanted: Robot-Dog Walkers. $20/Hour

https://www.nytimes.com/2026/10/27/technology/robot-dog-walkers.html
2•beansovertoast•18m ago•0 comments

Show HN: Search2o Simplifies AI Agents

https://search2o.com/
2•ramvenkat11•18m ago•0 comments

Show HN: Agent-relay for multi-vendor coding sessions

https://github.com/saranyan/agent-relay
2•svts•21m ago•0 comments

Bunkering Down

https://interconnected.org/home/2026/10/24/bunker
4•tonyedgecombe•22m ago•0 comments

Almost isn't good enough: Introducing VybBench, our Design Fidelity Benchmark

https://cocolabs.ai/blog/introducing-vybbench
2•dzats•22m ago•1 comments

Show HN: Qev – quake and local decision models

https://twitter.com/shinyrobject/status/2115124326194393595
2•kernel_sanders•22m ago•0 comments

iPhone Duo Breaks After 75,001 Folds Considerably Lower Than Samsung Foldables

https://gizmodo.com/iphone-duo-folding-test-worse-than-samsung-foldables-2000831131
3•HieronymusBosch•23m ago•0 comments
Open in hackernews

I built MCP on Ruby to help developers turn any Rails API into an MCP server

https://github.com/nagstler/mcp_on_ruby
2•NinjaGems•1y ago

Comments

NinjaGems•1y ago
I built MCP on Ruby, a gem that turns your Rails app into a fully-featured LLM server following the Model Context Protocol standard.

- What is it? Think of MCP as "REST for LLMs" - it standardizes how apps talk to AI models.

My implementation brings this to Ruby/Rails with:

Provider adapters for OpenAI & Anthropic (just add your API key)

Persistent storage options (memory, Redis, ActiveRecord)

Streaming responses for dynamic UIs

File handling & tool calling support

*Rails integration with just a few lines of code

- Why I built it I wanted a clean, Rails-friendly way to add AI capabilities without writing boilerplate for each provider. The existing MCP implementations were Python-focused, so I built this for the Ruby community.

- The ActiveRecord storage (just released in v0.3.0) lets you store conversations in your existing Rails database.

- Try it out: https://github.com/nagstler/mcp_on_ruby