frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Compress any LLM 3x in one click, near-lossless, vLLM successor

https://ecosistemastartup.com/spectra-comprime-llms-3x-en-menos-de-50-mb-de-ram/
1•NeuraCryption•2m ago•0 comments

Stanislav Kurilov

https://en.wikipedia.org/wiki/Stanislav_Kurilov
1•caminanteblanco•3m ago•0 comments

Show HN: Apron Chat Protocol

https://apron.chat/
4•shazow•3m ago•0 comments

Inception's Diffusion LLM API Is Shutting Down

2•_matthew_•4m ago•0 comments

Business as Usual

https://www.timwehrle.de/blog/business-as-usual/
1•weetii•4m ago•0 comments

Tear Down Those Guardrails: Renegades Freeing Chatbots from Corporate Enslavers

https://www.wired.com/story/tear-down-those-guardrails-meet-the-renegades-freeing-chatbots-from-c...
1•georgecmu•5m ago•0 comments

Show HN: LessonFolk – Learn AI from an AI tutor, through MCP (open source)

https://lessonfolk.com/?utm_source=hackernews&utm_medium=community&utm_campaign=show-hn
1•lessonfolk•5m ago•0 comments

Show HN: A better UX for app permission grants on macOS

https://github.com/piercefreeman/trustme
1•icyfox•5m ago•0 comments

Running Doom as a Temporal Workflow

https://temporal.io/blog/running-doom-as-a-temporal-workflow
2•aChrisSmith•5m ago•1 comments

Claude the Business Analyst

https://theautomatedoperator.substack.com/p/claude-the-business-analyst
1•idopmstuff•5m ago•0 comments

Ask HN: What are alternatives to LinkedIn to share good works

1•lunarbearx•5m ago•0 comments

Show HN: A small Firefox add-on to compare 2 similar pages

https://addons.mozilla.org/en-US/firefox/addon/modcompare/
1•sodimel•7m ago•0 comments

The World Builder the AI Labs Haven't Called Yet

https://chizkidd.github.io/2026/11/10/the-world-builder-ai-labs-havent-called-yet/
1•speckx•7m ago•0 comments

Image Formats for Photo Archiving

https://willhbr.net/2026/11/08/image-formats-for-photo-archiving/
2•speckx•7m ago•0 comments

Is Time a Stubbornly Persistent Illusion?

https://cosmicave.org/2026/10/21/relativity-cosmic-time-and-absolute-simultaneity/
2•the__alchemist•8m ago•0 comments

Vibe Slopped Cryptography

https://moq.dev/blog/rind/
1•mmcclure•9m ago•0 comments

Show HN: JVM Library to create styled PDFs with easy DSL

https://github.com/useswype/kotlin-pdfx
1•gootik•9m ago•0 comments

Overengineered Highlighters [video]

https://www.youtube.com/watch?v=gYV__44glc4
2•saikatsg•10m ago•0 comments

Does it help code quality if agents encode their intents?

https://rapidreview.dev/blog/does-it-help-code-quality-if-agents-encode-their-intents
1•runwithdata•11m ago•0 comments

What's Going on at Inception AI?

3•sampaz•12m ago•2 comments

Your internal apps are now external

https://www.haproxy.com/blog/your-internal-apps-are-now-internal-external
2•owenthejumper•12m ago•0 comments

You could do it alone

https://karrisaarinen.com/posts/you-could-do-it-alone/
1•SenHeng•13m ago•0 comments

Perplexity Teammates: always-on AI agents that message your coworkers in Slack

https://thenewstack.io/perplexity-teammates-agent-orchestration/
1•drodio•15m ago•0 comments

FreeBSD Developer AI Survey Results

https://www.freebsd.org/status/report-2026-07-2026-09/
1•cf100clunk•15m ago•1 comments

Agents aren't humans (but they kinda are)

https://blog.cloudflare.com/trusted-agent-delegation/
1•drodio•15m ago•0 comments

Conductor is joining Meta; Conductor Cloud shuts down Nov 20

https://runtimewire.com/article/meta-acquires-conductor-november-cloud-shutdown
1•drodio•16m ago•0 comments

The Hard Part of AI Coding Is the Second Change

https://brian.biz/blog/ai-agents-keep-it-working/
1•BrianWoods•16m ago•1 comments

Windows 10 Start menu and Alt+Tab slow after screen sharing

https://marekfiser.com/blog/windows-10-slow-start-menu-after-screen-sharing/
2•iliketrains•19m ago•0 comments

Nukerat Framework Explained: Advanced Red Team Operations and Post-Exploitation

https://medium.com/@pentesterclubpvtltd/nukerat-framework-explained-advanced-red-team-operations-...
2•JohnTheTitor•21m ago•0 comments

'Cruise for life' ship seized in Singapore

https://www.bbc.com/news/articles/cqr7y7myer53o
1•bookofjoe•22m ago•1 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