Hacker News

Article URL: https://timmb.com/when-you-cant-escape-an-interface/

Comments URL: https://news.ycombinator.com/item?id=47194319

Points: 1

# Comments: 0

Dedicated framework for building and orchestrating autonomous AI agent teams. Includes GitHub issue routing, multi-agent Discord coordination, dev pipeline orchestration, and complete documentation.

Downloadable guide: https://caffaro.gumroad.com/l/beqtp (PWYW, suggested $5)

Comments URL: https://news.ycombinator.com/item?id=47194313

Points: 1

# Comments: 0

Article URL: https://www.periodicspiral.com

Comments URL: https://news.ycombinator.com/item?id=47194305

Points: 1

# Comments: 0

We’ve been building a small design + navigation planning tool(https://no-edit.lovable.app) over the last few months, and while testing it, we ended up manually analyzing 312 landing pages across SaaS, indie projects, and AI tools. The original goal was simple: understand how people structure navigation before designing UI. What we found surprised us. 1. 68% of pages had unclear primary navigation Either: Too many items (7+ top-level links) Or vague labels like “Solutions”, “Platform”, “Explore” Users had to think before clicking. 2. Most CTAs compete with each other In 54% of cases: 2–3 primary buttons had equal visual weight. No obvious action hierarchy. Design looked clean. Decision-making wasn’t. 3. Mobile navigation is often an afterthought A lot of responsive menus technically “worked”, but: Important links were buried CTA visibility dropped significantly 4. Information hierarchy ≠ visual hierarchy Several pages looked polished (good colors, spacing, typography), but: Navigation structure didn’t reflect user journey. Sections were ordered for storytelling, not usability. The interesting part: When we restructured some of these flows into simple sitemap-style diagrams first, clarity improved immediately — even before touching UI. It made me think that most tools focus heavily on design layers (fonts, templates, components), but skip structured navigation thinking early on. I’m curious: Do you plan navigation before visual design? Or do you design first and adjust structure later? Would love feedback from people who’ve built and tested landing pages at scale.

Comments URL: https://news.ycombinator.com/item?id=47194296

Points: 1

# Comments: 1

Article URL: https://aeon.co/essays/psychopathy-is-a-zombie-idea-why-does-it-cling-on

Comments URL: https://news.ycombinator.com/item?id=47194275

Points: 1

# Comments: 0

Article URL: https://twitter.com/trbrtc/status/2027703248887427576

Comments URL: https://news.ycombinator.com/item?id=47194238

Points: 1

# Comments: 0

I built this because prompts kept ending up in the worst possible places — Python strings, Notion docs, `.txt` files, Slack threads. There was no clean way to version them, diff them, or test the same prompt across different models without writing a throwaway script.

prompt-run treats `.prompt` files as first-class runnable artifacts. A `.prompt` file is a YAML header (model, provider, temperature, variable declarations) followed by a plain text body with `{{variable}}` substitution. You run it from the terminal:

``` prompt run summarize.prompt --var text="$(cat article.txt)" ```

You can override model and provider at runtime without editing the file:

``` prompt run summarize.prompt --model gpt-4o --provider openai ```

The `prompt diff` command runs the same prompt for two different inputs (or two prompt versions against the same input) and shows outputs side by side. That's the feature I find most useful when iterating.

Supports Anthropic, OpenAI, and Ollama out of the box. MIT license. No telemetry, no accounts, no backend — just a local CLI tool that talks directly to whichever provider you configure.

The file lives in your repo, gets versioned by git, and can be reviewed in PRs like any other code.

Would be curious to hear whether others have hit this same friction and how you've handled it.

PyPI: https://pypi.org/project/prompt-run/

Comments URL: https://news.ycombinator.com/item?id=47194215

Points: 1

# Comments: 0

Article URL: https://github.com/CodeZeno/Claude-Code-Usage-Monitor

Comments URL: https://news.ycombinator.com/item?id=47194211

Points: 1

# Comments: 1

Article URL: https://krebsonsecurity.com/2026/02/who-is-the-kimwolf-botmaster-dort/

Comments URL: https://news.ycombinator.com/item?id=47194206

Points: 1

# Comments: 0

Hi HN,

I’ve been building Opencom, an open-source customer messaging platform - basically an Intercom alternative you can run yourself with minimal setup or infrastructure to manage. Repo + demo here:

https://github.com/opencom-org/opencom

https://opencom.dev

I started working on it after years of frustration using Intercom. I love the selection of tools they offer, but not their complex and expensive pricing model or their frequent layout and UX changes. I wanted an open source alternative that was simple and cheap to run, making it accessible for projects of any size. I wanted something that was as close to Intercom's feature-set as possible, and then some. You can self host as much or as little as you like, from everything to nothing, using just the hosted frontends and backend.

What it includes

- Real-time chat via an embeddable widget, connected to a shared inbox - Tickets, campaigns/series, surveys (NPS/CSAT), segments, reports - Product tours/tooltips/checklists - Knowledge base + an AI agent with human handoff - Email channel via Resend integration - React Native SDK - with native iOS and Android SDKs planned

Hosting / deployment model

Opencom is set up so you can keep the architecture fairly simple: Backend: everything runs on Convex (the repo’s deploy guide treats Convex as the canonical backend), and email features use Resend (e.g., RESEND_API_KEY, webhook verification, etc.). Frontend: the web dashboard/widget/landing are standard Next.js/Vite apps, so you can host them on Vercel if you like (Convex has first-class docs/integration for deploying alongside Vercel).

Mix-and-match: there’s an explicit “hosted apps + your own Convex backend” profile, as well as “self-host web + your own backend” and “full self-host”.

In other words, you can either host everything yourself, or use the hosted web + mobile apps while pointing them at your Convex backend.

React Native SDK

If you want Intercom-style in-app messaging inside your own mobile app, Opencom includes a React Native SDK package in the monorepo, alongside native iOS/Android SDKs planned.

About the hosted offering

At the moment, the hosted setup is intended as a demo (so people can kick the tyres without doing a full deploy). If there’s enough demand for a reasonably priced, direct Intercom alternative, I could shift it into a proper hosted service.

Feedback I’d really appreciate

- Does the Convex-backend + Vercel-frontend split feel sensible for teams who’d otherwise buy Intercom? - Anything missing that’s a hard requirement for switching? - Thoughts on licensing / pricing expectations if this became a straightforward hosted option (no enterprise sales theatrics)? - Any other thoughts - this is my first open sourced project, and it is built heavily with AI, so I expect there to be issues / silly mistakes, and I know there are areas that need work. My hope is that others want a product like this and would be open to chipping in / fleshing out features they want or need.

Thanks for checking it out!

Comments URL: https://news.ycombinator.com/item?id=47194196

Points: 1

# Comments: 0

Article URL: https://www.lenna.org/lenna_visit.html

Comments URL: https://news.ycombinator.com/item?id=47194174

Points: 1

# Comments: 1

Article URL: https://developers.cloudflare.com/changelog/post/2026-02-26-async-stale-while-revalidate/

Comments URL: https://news.ycombinator.com/item?id=47194160

Points: 2

# Comments: 0

Article URL: https://apnews.com/live/live-updates-israel-iran-february-28-2026

Comments URL: https://news.ycombinator.com/item?id=47194147

Points: 1

# Comments: 0

Article URL: https://www.theguardian.com/science/2026/feb/27/nasa-changes-delays-moon-missions

Comments URL: https://news.ycombinator.com/item?id=47194144

Points: 1

# Comments: 0

Article URL: https://www.nytimes.com/2026/02/28/world/middleeast/iran-attacks-what-to-know.html

Comments URL: https://news.ycombinator.com/item?id=47194132

Points: 1

# Comments: 1

Article URL: https://techcrunch.com/2026/02/22/move-over-apple-meet-the-alternative-app-stores-available-in-the-eu-and-elsewhere/

Comments URL: https://news.ycombinator.com/item?id=47194119

Points: 1

# Comments: 0

Article URL: https://www.trustedreviews.com/how-to/remove-chatgpt-apple-intelligence-4580438

Comments URL: https://news.ycombinator.com/item?id=47194087

Points: 2

# Comments: 0

Article URL: https://github.com/mubit-ai/codaph

Comments URL: https://news.ycombinator.com/item?id=47194077

Points: 1

# Comments: 0

Article URL: https://www.newyorker.com/magazine/1994/07/25/how-i-met-my-wife

Comments URL: https://news.ycombinator.com/item?id=47194050

Points: 1

# Comments: 0

Code, as such, is a detailed, verifiable spec that a machine can execute. LLMs are already great at translating code to natural language and vice-versa. Why do we need a second, less detailed and less verifiable copy of the code?

Comments URL: https://news.ycombinator.com/item?id=47194035

Points: 1

# Comments: 5