Hacker News

Article URL: https://www.xda-developers.com/sudo-rs-password-asterisk-entry/

Comments URL: https://news.ycombinator.com/item?id=47177851

Points: 1

# Comments: 0

Article URL: https://takeinterest.ai/guardclaw

Comments URL: https://news.ycombinator.com/item?id=47177832

Points: 1

# Comments: 1

I kept running into the same problem: trying to find something new to watch, opening multiple streaming apps, and Googling titles one by one because the apps don't tell you if something is actually worth watching.

So I built What to Watch, which aggregates content across major streaming services and reduces discovery to a simple Watch or Skip decision using IMDb and community ratings.

I hadn’t written code since high school — this was built over ~400 iterations using Claude and Replit.

Still very early and mostly trying to learn how people actually decide what to watch and how broad this use case is. Would love feedback from the HN community.

Comments URL: https://news.ycombinator.com/item?id=47177814

Points: 1

# Comments: 1

Article URL: https://atoship.com/docs/integrations/openclaw

Comments URL: https://news.ycombinator.com/item?id=47177794

Points: 1

# Comments: 0

Article URL: https://pudding.cool/2026/02/happy-map/

Comments URL: https://news.ycombinator.com/item?id=47177788

Points: 1

# Comments: 0

Article URL: https://www.cnn.com/2026/02/26/business/block-layoffs-ai-jack-dorsey

Comments URL: https://news.ycombinator.com/item?id=47177785

Points: 1

# Comments: 0

Article URL: https://mailvaultapp.com/

Comments URL: https://news.ycombinator.com/item?id=47177782

Points: 2

# Comments: 1

Hi all,

We’ve been thinking about a core limitation in current mobile AI assistants:

Most systems (e.g., Apple Intelligence, Google Assistant–style integrations) rely on predefined schemas and coordinated APIs. Apps must explicitly implement the assistant’s specification. This limits extensibility and makes the ecosystem tightly controlled.

On the other hand, GUI-based agents (e.g., AppAgent, AutoDroid, droidrun) rely on screenshots + accessibility, which gives broad power but weak capability boundaries.

So we built Mobile-MCP, an Android-native realization of the Model Context Protocol (MCP) using the Intent framework.

The key idea:

- Apps declare MCP-style capabilities (with natural-language descriptions) in their manifest.

- An LLM-based assistant can autonomously discover all exposed capabilities on-device via the PackageManager.

- The LLM selects which API to call and generates parameters based on natural language description.

- Invocation happens through standard Android service binding / Intents.

Unlike Apple/Android-style coordinated integrations:

- No predefined action domains.

- No centralized schema per assistant.

- No per-assistant custom integration required.

- Tools can be dynamically added and evolve independently.

The assistant doesn’t need prior knowledge of specific apps — it discovers and reasons over capabilities at runtime.

We’ve built a working prototype + released the spec and demo:

GitHub: https://github.com/system-pclub/mobile-mcp

Spec: https://github.com/system-pclub/mobile-mcp/blob/main/spec/mobile-mcp_spec_v1.md

Demo: https://www.youtube.com/watch?v=Bc2LG3sR1NY&feature=youtu.be

Paper: https://github.com/system-pclub/mobile-mcp/blob/main/paper/mobile_mcp.pdf

Curious what people think:

Is OS-native capability broadcasting + LLM reasoning a more scalable path than fixed assistant schemas or GUI automation?

Would love feedback from folks working on mobile agents, security, MCP tooling, or Android system design.

Comments URL: https://news.ycombinator.com/item?id=47177770

Points: 1

# Comments: 0

Article URL: https://www.reuters.com/business/blocks-fourth-quarter-profit-rises-announces-over-4000-job-cuts-2026-02-26/

Comments URL: https://news.ycombinator.com/item?id=47177733

Points: 1

# Comments: 0

Article URL: https://finditsales.substack.com/p/why-the-storytellers-will-win-the

Comments URL: https://news.ycombinator.com/item?id=47177732

Points: 1

# Comments: 0

Article URL: https://github.com/openoctp/spec

Comments URL: https://news.ycombinator.com/item?id=47177702

Points: 1

# Comments: 1

Article URL: https://evanhahn.com/introducing-gzpeek/

Comments URL: https://news.ycombinator.com/item?id=47177700

Points: 1

# Comments: 0

Article URL: https://larahogan.me/blog/creating-momentum-when-an-employee-is-stuck/

Comments URL: https://news.ycombinator.com/item?id=47177698

Points: 1

# Comments: 0

Article URL: https://sny.sh/hypha/blog/scp

Comments URL: https://news.ycombinator.com/item?id=47177675

Points: 1

# Comments: 0

Article URL: https://unfoldingdiagrams.leaflet.pub/3mft6olldos26

Comments URL: https://news.ycombinator.com/item?id=47177674

Points: 2

# Comments: 0

Article URL: https://architectelevator.com/strategy/economics-technical-speaking/

Comments URL: https://news.ycombinator.com/item?id=47177668

Points: 1

# Comments: 0

Article URL: https://johnnysswlab.com/exposing-more-parallelism-is-the-hidden-reason-why-some-vectorized-loops-are-faster-not-vectorization-per-se/

Comments URL: https://news.ycombinator.com/item?id=47177651

Points: 2

# Comments: 0

Article URL: https://www.araju.dev/blog/ai-for-social-impact

Comments URL: https://news.ycombinator.com/item?id=47177635

Points: 1

# Comments: 0

Hi HN, I’m the creator of ORBTX.

I built this because most tools in astrodynamics are either overly simplistic web calculators or heavyweight, expensive legacy suites that look like they were made in the 90s. I wanted to create a "middle ground": professional-grade orbital computation that is fast, visual, and API-first.

The Tech Stack:

Frontend: Next.js & React.

3D Engine: Three.js (React Three Fiber) for real-time trajectory rendering.

Physics: Custom engine handling classical two-body mechanics (Hohmann and Bi-elliptic transfers).

Key Challenges: One of the main hurdles was maintaining floating-point precision for astronomical distances within a browser environment while keeping the 3D visualization smooth at 60fps. I’m currently refining the propagator to handle more complex perturbations in future updates.

The Goal: To move orbital mechanics away from "Excel-engineering" into a modern, developer-ready infrastructure. Think of it as a step toward "Figma for orbital mechanics."

No sign-up required, it's open for testing. I’d love to hear your thoughts on the physics implementation, the API structure, or any edge cases you find!

Comments URL: https://news.ycombinator.com/item?id=47177432

Points: 1

# Comments: 0

Article URL: https://pydantic.dev/articles/the-human-in-the-loop-is-tired

Comments URL: https://news.ycombinator.com/item?id=47177411

Points: 2

# Comments: 0