frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Texas leads nation in proposed power plants for data centers

https://www.texastribune.org/2026/07/01/texas-data-center-power-plans-emissions/
1•measurablefunc•3m ago•0 comments

Show HN: Tag: Open-source NVMe cache for object storage

1•xena•3m ago•0 comments

Parisian Prisoners Offered Freedom If They Married Prostitutes Moved to Louisana

https://historycollection.com/parisian-prisoners-offered-freedom-agreed-marry-prostitutes-move-mi...
1•arealaccount•3m ago•0 comments

Banning AI in Law School: We've Seen This Before

https://hardcoresoftware.learningbyshipping.com/p/240-banning-ai-in-law-school-weve
1•bookofjoe•3m ago•0 comments

Everybody Should Welcome Nationalizing AI

https://jacobin.com/2026/07/ai-nationalization-sanders-libertarians-property
1•robtherobber•4m ago•0 comments

Show HN: Running over 80M tokens in one agent session with no compaction

https://github.com/Kiz8-Team/pi-cwl
1•andrew_ocs•4m ago•0 comments

Touchscreens at Risk: A Practical Side-Channel Phone Attack via EM Emanations

https://arxiv.org/abs/2605.14633
1•CGMthrowaway•5m ago•0 comments

GraalJS 25.1: Modern JavaScript on GraalVM

https://medium.com/graalvm/graaljs-25-1-modern-javascript-on-graalvm-25f5c7cea361
1•grashalm•6m ago•0 comments

Show HN: Choji – turn Slack threads into ready-to-merge PR with Choji

https://choji.ai/agents/slack
1•chptung•7m ago•0 comments

Chai Discovery, an A.I. Drug Startup, Raises $400M

https://www.nytimes.com/2026/07/14/business/dealbook/chai-discovery-ai-drug-development.html
2•brandonb•8m ago•0 comments

Show HN: Like Audiosurf but writen in Odin and open source

https://sekor.eu.org/psycho-odin/
1•modinfo•8m ago•0 comments

VAID – proof-of-possession signing for AI agents, no network call to verify

https://github.com/solara-associates/vaid
1•solara123•8m ago•0 comments

Commission adds exemptions to rules on removability of portable batteries

https://environment.ec.europa.eu/news/commission-adds-exemptions-portable-battery-removal-rules-2...
2•hn_acker•9m ago•0 comments

Building Maximally Efficient Cloud Environments for Long-Horizon Agents

https://www.sailresearch.com/blog/maximally-efficient-cloud-environments
1•blintz•9m ago•0 comments

Show HN: BeePub – A reading-first alternative to Calibre-Web

https://github.com/oalieno/beepub
2•sheldon7337•10m ago•0 comments

Searching for the Best Silicone USB Cable

https://www.frankchiarulli.com/blog/best-silicone-usb-cable/
1•fcjr•11m ago•0 comments

A Cross-Platform Rust UI Framework with Qt

https://www.qt.io/blog/rust-ui-framework-via-bridging-technology
1•coffeeaddict1•11m ago•0 comments

The Bug Was in My Beliefs, Not My Code

https://medium.com/@alanscottencinas/the-bug-was-in-my-beliefs-not-my-code-f6bb717f952e
1•encinas88•11m ago•0 comments

Pseudpocalypse: Conjecture: To interact with the world is to identify yourself

https://dynomight.substack.com/p/pseudpocalypse
1•crescit_eundo•12m ago•0 comments

Pollution from Musk's unpermitted xAI power project hits in Black communities

https://www.reuters.com/world/americas/pollution-musks-unpermitted-xai-power-project-hits-hardest...
9•onemoresoop•12m ago•0 comments

The Tao Te Ching: The Ancient Case for Letting Go

https://yalebooks.yale.edu/2026/06/18/the-tao-te-ching-the-ancient-case-for-letting-go/
2•amemi•13m ago•1 comments

AI Took my coding. Whats left for me?

https://loopholelabs.io/blog/ai-took-my-coding-whats-left-for-me#zx81--qy10
1•axod•13m ago•0 comments

Suddenly, Hormuz Is Less Crucial Than It Was

https://paulkrugman.substack.com/p/suddenly-hormuz-is-less-crucial-than
2•SubiculumCode•13m ago•1 comments

ChatGPT Mac App ruins Chats interface by merging with Codex

https://chatgpt.com/download/
2•midnight_eclair•15m ago•1 comments

Let's build a children's public internet

https://www.theverge.com/report/962823/childrens-public-internet-child-safety-proposal
1•Brajeshwar•16m ago•0 comments

Show HN: AI-powered SEO automation for WordPress

https://saas.seamanshelp.com/
1•not_wowinter13•16m ago•0 comments

Show HN: GLP-RAM – Chrome extension reducing the browser memory appetite

https://github.com/chebykinn/glp-ram
1•mifydev•16m ago•0 comments

Integrating .NET GC in your C++ application

https://kant2002.github.io/en/dotnet/open-source/2026/07/12/dotnet-gc.html
1•kant2002•17m ago•1 comments

Show HN: Vibe-coding games with Claude, day 93: Rock Paper Scissors Lizard Spock

https://gamevibe.us/93-rock-paper-scissors-lizard-spock
2•pzxc•18m ago•0 comments

Cloudflare is now "Local First, Everywhere"

https://bsky.app/profile/jakelazaroff.com/post/3mqlpvpjswk2k
1•embedding-shape•19m ago•0 comments
Open in hackernews

Show HN: Skills as a Service via MCP –> Coding Agent Skill Library

https://github.com/mmccalla/coding-agent-skill-library
1•get-analyst-01•1h ago
Hi HN!,

While discussing agentic engineering with my close network, it came to light that most agent skills still live on desktops, in repositories, or in people’s heads.

In the context of large, regulated organisations, where compliance and governance are key drivers, it makes sense for skills to be treated like any other governed, trusted and compliant data asset.

From a software and agentic engineering perspective, the tooling should be easy to adopt and make agentic coding "easier, faster, safer".

This started to shape a new capability, “Skills as a Service”, delivered by the Coding Agent Skill Library, which moves skills into a secure, governed and trusted service.

It has two flavours:

1. Drop-in library: If you don't want to change your current workflow, pull any skills you need directly into your environment and use them with Claude, Codex, Cursor, Antigravity, and other agent frameworks or IDEs.

2. Service via MCP: If you want Skills as a Service, then Agents and IDEs access a semantically linked, ontology-first knowledge graph through a read-only MCP server. This is where the real value is.

The knowledge graph is designed to retrieve the right skills for the shape of the task, rather than relying only on names, folders or basic metadata. It models task intent, workflow stage, supporting evidence, skill versions and relationships between skills.

Runs locally in docker, wired into VSCode or Cursor using stdio. Intent is this would be scaled out for enterprise environments, have inner source style contributions federated ops model. My thinking was to enable controlled contribution, consistent quality, compliance and simplify skill discovery and reduce drift, by having a single trusted source.

The aim was simple: make curated skills a safe, versioned and governed organisational asset, accessible from a single internal service. Question is does it actually help make real builders d2d - faster, safer, cheaper, easier and does it help enterprises govern skills usage?

Repo: https://github.com/mmccalla/coding-agent-skill-library