Hacker News

Article URL: https://tminus.one/

Comments URL: https://news.ycombinator.com/item?id=48199598

Points: 2

# Comments: 2

Article URL: https://pinionengineering.github.io/storage-proofs/

Comments URL: https://news.ycombinator.com/item?id=48199586

Points: 1

# Comments: 1

Hi HN!

I've built a plugin for Obsidian that adds lightweight, highly customized, and native-feeling chart layouts (Bar, Line, Pie) specifically for the new database-like "Bases" feature.

Repository: https://github.com/zobweyt/obsidian-bases-chart-layouts Plugin page: https://community.obsidian.md/plugins/bases-chart-layouts

### The Problem

I needed a way to visualize my structured note data in Obsidian without the bloat. Existing chart solutions either required writing complex JavaScript inside codeblocks or felt like heavy web-views slapped onto the app.

I wanted charts that feel like a built-in, core feature of Obsidian—lightweight, snappy, and deeply integrated with the UI.

### What makes it "Native-Feeling"

It goes beyond basic SVG rendering to mimic Obsidian’s native desktop UX patterns:

- *Interactions:* Full support for `cmd/ctrl + click` and middle-click (`auxclick`) to open underlying data files in new tabs or split panes. - *Context Menus:* Right-clicking the chart triggers native Obsidian context menus, not generic browser ones. - *Theme Adaptation:* It doesn't just switch light/dark modes. It dynamically inherits your exact Obsidian theme CSS variables and fonts.

### Tech Stack & DX

I wanted to keep the codebase as clean as the UI:

* *No Node/NPM Boilerplate:* The entire project is built purely with Deno (fmt, lint, build). This kept the development environment exceptionally lightweight and fast. * *Rendering:* Powered by Apache ECharts under the hood for crisp vector rendering. The final plugin size is only ~1MB. * *I18n:* Built-in localization support (currently English and Russian).

### How to try it

1. Ensure you have Bases enabled in Obsidian. 2. Add a chart layout to your view. 3. Map your columns using the native Properties panel.

I’d love to hear your thoughts on the UX integration, and especially on using Deno for building Obsidian plugins. Any feedback or feature requests are welcome!

Comments URL: https://news.ycombinator.com/item?id=48199577

Points: 1

# Comments: 0

Article URL: https://twitter.com/addictedtoigno1/status/2056072770987970825

Comments URL: https://news.ycombinator.com/item?id=48199575

Points: 1

# Comments: 1

Hi HN,

I built Limitless because I wanted to create a unified, high-fidelity platform for Open Source Intelligence (OSINT), dark web collection, and interactive threat training.

Most cybersecurity training systems are either standard slide decks or text-heavy questionnaires. At the same time, actual analyst tools are scattered across hundreds of disconnected command-line scripts and API wrappers.

Limitless merges real analyst tools with a reactive, browser-based training simulator.

The Ecosystem Components: 1. *Feynman (OSINT search): An intelligence search engine that maps digital footprints, uncovers hidden connections, and aggregates intelligence across 200+ online sources instantly. 2. *Sentinel (Darknet Agent):* An autonomous AI agent designed for deep-web collection, learning, and reasoning across darknet forums. 3. *Interactive Training Sandboxes:* Scenario-driven tutorials that teach cybersecurity operations through visual tools: * GEOINT Simulator: Uses Leaflet and coordinates to calculate physical proximity (Haversine formula), supporting tolerance radii and partial scoring. * Steganography Lab: Toggles Red/Green/Blue color channels and bitplane depths on canvas elements dynamically in the browser. * Draggable Chronology & Classification: Drags event timelines and classifies logs using `@dnd-kit`. * Audio & Video Intel: Custom playback speed modulators and zoom canvas overlays.

Localization: Reactive, real-time translations supporting 9 languages (English, Spanish, Portuguese, French, Russian, Hebrew, Arabic, Japanese, and Hindi).

You can explore the interactive intelligence based sandboxes for learning directly https://limitless-osint.com/

I’d love to hear your feedback on the investigative tools (Feynman & Sentinel), the workspace user experience, or what intelligence tools you would like to see us build next.

Thanks!

Comments URL: https://news.ycombinator.com/item?id=48199574

Points: 1

# Comments: 0

Article URL: https://gisbi.me/blog/article-1779175137638

Comments URL: https://news.ycombinator.com/item?id=48199572

Points: 1

# Comments: 0

Article URL: https://thevarsity.ca/2026/03/15/the-enhanced-games-the-line-we-keep-moving/

Comments URL: https://news.ycombinator.com/item?id=48199554

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=SbBnc8FjUXM

Comments URL: https://news.ycombinator.com/item?id=48199512

Points: 1

# Comments: 0

Article URL: https://www.dwarkesh.com/p/the-mistake-of-conflating-intelligence

Comments URL: https://news.ycombinator.com/item?id=48199479

Points: 1

# Comments: 0

Article URL: https://blog.quent.in/blog/2026/05/19/your-small-pr-rule-wont-survive-ai/

Comments URL: https://news.ycombinator.com/item?id=48198950

Points: 1

# Comments: 0

Article URL: https://github.com/vercel-labs/zero

Comments URL: https://news.ycombinator.com/item?id=48198944

Points: 1

# Comments: 0

Article URL: https://fprox.substack.com/p/risc-v-and-floating-point

Comments URL: https://news.ycombinator.com/item?id=48198917

Points: 1

# Comments: 0

Article URL: https://andreasflakstad.no/posts/kimen/

Comments URL: https://news.ycombinator.com/item?id=48198916

Points: 1

# Comments: 0

Article URL: https://www.amd.com/en/resources/product-security/bulletin/amd-sb-7052.html

Comments URL: https://news.ycombinator.com/item?id=48198896

Points: 1

# Comments: 0

Clicking on some FAQs (like "Does grpcurl Does grpcurl require a .proto file?a .proto file?") redirects me to https://i3seperfiles.com a really suspicious web site). Am I crazy or over-paranoid?

Update: nevermind that's been discussed here https://www.fullstory.com/blog/inside-a-global-campaign-hijacking-open-source-project-identities/ Sad to see an Open source project targeted like that

Comments URL: https://news.ycombinator.com/item?id=48198841

Points: 2

# Comments: 0

Article URL: https://arstechnica.com/gadgets/2026/05/plexs-200-lifetime-pass-price-hike-tries-forcing-users-to-another-subscription/

Comments URL: https://news.ycombinator.com/item?id=48198836

Points: 2

# Comments: 0

Built Bevel because my girlfriend is in a book club and I wanted to make her something. It’s a daily puzzle: you read a 200ish-word public-domain passage and try to deduce the author, decade, nationality, and title in five guesses with adaptive hints in between guesses. Wordle-ish mechanics, but for classic literature. There’s a Discord activity too if you enjoy it & want to add it to your server. Bevel is currently unverified on Discord, so only addable to smaller servers for now. Would love any feedback.

Comments URL: https://news.ycombinator.com/item?id=48198827

Points: 2

# Comments: 0

I was job hunting across countries and kept running into the same two problems.

First: every platform optimises for different things. LinkedIn wants keywords. ATS parsers want specific formatting. GitHub rewards README quality. A portfolio has to be crawlable. Staying consistent across all of them without a system is nearly impossible.

Second: every AI agent session starts from scratch. Paste a context dump, watch it burn half the window fetching your links, get generic output. The agent has no grounded facts about you – so it improvises.

Both problems point to the same fix: one verified source of truth, written once, readable by humans and agents alike.

That's AgentKit SEO [1]. The core concept is an agent-context file – a private Markdown document with your career facts, project metrics, and proof links. You build it once using the context skill. Every other skill (LinkedIn, GitHub, CV/ATS, portfolio, X) reads from it.

It ships as a CLI that installs focused skill files into whichever agent you use:

npx agentkit-seo install --provider codex # or: claude-code, gemini-cli, opencode, or portable SKILL.md This is not an auto-apply tool. It's the step before that: create one grounded file, then use it as input for any agent, any platform, any task. No hallucination from context overload. No contradictions between your CV and your LinkedIn.

Open source, free, MIT licence. My co-author Elia Innocenti and I built it because it didn't exist.

[1] https://github.com/agentkit-seo/agentkit-seo

[2] https://agentkit-seo.github.io/

[3] https://www.npmjs.com/package/agentkit-seo

Comments URL: https://news.ycombinator.com/item?id=48198809

Points: 1

# Comments: 0

I have a long YouTube watch-later playlist and wanted Claude to help me triage it by summarizing each video. It couldn't. I later discovered YouTube is not accessible to Claude, ChatGPT and majority of AIs. So I built this MCP server to fix that.

This is a local MCP server that gives any MCP-compatible client (Claude, ChatGPT, OpenClaw and others) 8 tools for working with YouTube content:

- fetch full transcript, with or without timestamps - download transcript as a .md file - get video metadata (title, channel, views, duration, likes, description) - search within captions for a keyword or phrase - download video or audio

No YouTube API key. No account. No signup. Everything runs locally on your machine against publicly available data.

Setup is one line: npx @umbertotancorre/youtube-mcp

yt-dlp and ffmpeg are bundled automatically during install, so there's nothing to configure manually.

It's free and MIT licensed. Feedbacks are welcome.

Comments URL: https://news.ycombinator.com/item?id=48198790

Points: 3

# Comments: 0

Article URL: https://www.stopspying.org/biometric-map

Comments URL: https://news.ycombinator.com/item?id=48198777

Points: 2

# Comments: 0