Hacker News

Hey HN, we've been using Claude Code daily for months, and built Scape to fix our biggest pain point: managing multiple worktrees quickly & reducing the mental load of switching between them quickly.

What we built: Scape is a macOS menu bar app that sits on top of Claude Code and gives you:

- One-click worktrees: click a button, get a new git worktree with a Claude session already working on it. Develop in parallel without leaving your current branch. - Toolkit: per-session actions like "Create PR", "Commit & Push", "Run & Fix Tests." Add your own with bash scripts or prompts. - Orchestrators: auto-answer questions, auto-approve tools, set custom instructions. Sessions run autonomously while you work on something else.

It monitors all your Claude Code sessions across iTerm2 terminals, so you always know what's happening at a glance.

Privacy: Everything is stored locally at ~/.claude/scape/. No code or terminal content leaves your machine.

We'd love feedback, especially on what workflows you'd want to automate. We're planning to add A LOT more over the coming weeks, specifically around embedded terminals and more automation.

macOS 14+ only for now (more terminal & agent model support coming).

Comments URL: https://news.ycombinator.com/item?id=47257712

Points: 3

# Comments: 0

Article URL: https://www.youtube.com/watch?v=QFqsY-DT6rg

Comments URL: https://news.ycombinator.com/item?id=47257711

Points: 1

# Comments: 0

Article URL: https://github.com/ratwolfzero/Chladni_Figures

Comments URL: https://news.ycombinator.com/item?id=47257680

Points: 2

# Comments: 1

I built this because MCP supports remote servers over Streamable HTTP, but a lot of MCP servers still assume the client and server share a filesystem. Once the server is in a container/VM, two things break: tools can’t read client-local file paths, and outputs (screenshots/PDFs etc) get written on the server where the client can’t access them.

Remote MCP Adapter is an application-layer gateway that sits between an MCP client and one/more upstream MCP servers. It stages inputs as upload handles and captures outputs as MCP resources that the client can read back. It also adds session isolation, automated TTL-based cleanups, quotas, state persistence, health checks and the other standard constructs necessary for a production deployment.

Docs + demo: https://aakashh242.github.io/remote-mcp-adapter/ Code: https://github.com/aakashh242/remote-mcp-adapter

Quickstart is in the README (docker compose spins up Playwright MCP + the adapter) and you point your agent at: http://localhost:8932/mcp/playwright

Would love feedback on edge cases and missing features: artifact capture rules across different servers, guarantee that all requests with a given session_id are consistently routed to the same upstream.

Comments URL: https://news.ycombinator.com/item?id=47257661

Points: 1

# Comments: 0

Article URL: https://dylan.gr/1772520728

Comments URL: https://news.ycombinator.com/item?id=47257637

Points: 3

# Comments: 0

Article URL: https://www.propublica.org/article/nike-jobs-indonesia-living-wages

Comments URL: https://news.ycombinator.com/item?id=47257622

Points: 4

# Comments: 0

Article URL: https://duanehilton.com/notes/your-first-open-source-pr

Comments URL: https://news.ycombinator.com/item?id=47257607

Points: 1

# Comments: 0

Article URL: https://github.com/brandontan/signalbase

Comments URL: https://news.ycombinator.com/item?id=47257575

Points: 1

# Comments: 0

Article URL: https://github.com/octoflow-lang/octoflow

Comments URL: https://news.ycombinator.com/item?id=47257573

Points: 1

# Comments: 3

Article URL: https://dlants.me/ai-se.html

Comments URL: https://news.ycombinator.com/item?id=47257438

Points: 1

# Comments: 0

Article URL: https://www.cnbc.com/2026/03/04/amazon-bahrain-data-centers-targeted-iran-drone-strike.html

Comments URL: https://news.ycombinator.com/item?id=47257431

Points: 1

# Comments: 0

Article URL: https://www.timesofisrael.com/liveblog-march-05-2026/

Comments URL: https://news.ycombinator.com/item?id=47257427

Points: 1

# Comments: 0

Article URL: https://github.com/akshan-main/vibe-check/README.md

Comments URL: https://news.ycombinator.com/item?id=47257420

Points: 1

# Comments: 1

Article URL: https://www.bloomberg.com/news/articles/2026-03-05/anthropic-s-amodei-reopens-ai-discussions-with-pentagon-ft-says

Comments URL: https://news.ycombinator.com/item?id=47257403

Points: 1

# Comments: 0

Article URL: https://acko.net/blog/the-l-in-llm-stands-for-lying/

Comments URL: https://news.ycombinator.com/item?id=47257394

Points: 1

# Comments: 0

Hi HN, I’m a frontend engineer and built Jobbi.app, a small tool that uses AI to tailor a resume to each job description. It’s currently free, unlimited, with PDF export. Problem During my own job searches I kept rewriting the same resume for dozens of roles. Most “AI resume builders” I tried either: – focused on generating a resume from scratch instead of working with my existing one – had strict free‑tier limits or aggressive upsells – produced very generic, obviously‑AI text. What Jobbi does – you upload a “master” resume (PDF or text) – you paste a job description – it extracts relevant parts, rewrites bullets to match the JD, and outputs a tailored resume you can edit in the browser and export as PDF. Under the hood it: – parses the resume into sections (experience, skills, etc.) – scores which bullets/skills are relevant to the JD keywords – rewrites only those parts, keeping the original structure and tone as much as possible – uses Gemini API with some prompt engineering around “don’t invent experience / don’t change dates / keep numbers”. What’s different – it assumes you already have a resume and just want fast tailoring, not a full CV builder – no limits / logins / credit cards right now (I care more about usage + feedback than monetization at this stage) – I’m optimizing specifically for software/tech roles, so I’d love feedback from this crowd. Questions for you – For those who’ve used similar tools, what did you hate the most? – Is there something obviously broken / insecure / unethical in my approach? – What features would make this actually useful for HN readers (e.g. integration with “Who’s Hiring”, diff view of changes, local‑only mode, OSS version, etc.)? Link: https://jobbi.app I’ll be in the thread to answer questions and hear any feedback (incl. “don’t do this because X”).

Comments URL: https://news.ycombinator.com/item?id=47257386

Points: 1

# Comments: 0

I built Poppy as a side project to help people keep in touch more intentionally. Would love feedback on onboarding, reminders, and overall UX. Happy to answer questions.

Comments URL: https://news.ycombinator.com/item?id=47257352

Points: 1

# Comments: 0

Article URL: https://franken.style/

Comments URL: https://news.ycombinator.com/item?id=47257349

Points: 1

# Comments: 1

Article URL: https://arpitbhayani.me/blogs/bm25/

Comments URL: https://news.ycombinator.com/item?id=47257341

Points: 2

# Comments: 0

I'm about to start a degree in computer science + political science and need a laptop. I want something small and light but I'm poor. Should I get the new Macbook Neo or a second hand X1 carbon from 3-4 years ago.

Comments URL: https://news.ycombinator.com/item?id=47257320

Points: 1

# Comments: 1