The core loop: dump accessibility tree via uiautomator → parse and filter to ~40 relevant elements → LLM returns {think, plan, action} → execute via ADB → repeat.

Some technical decisions worth noting:

- Primary input is the accessibility tree, not vision. Vision (screenshots + multimodal model) is only a fallback for when the tree is empty (WebViews, Flutter).

- Stuck detection: if the screen state doesn't change for 3 steps, recovery kicks in with back navigation, home, or app re-launch.

- Two execution modes: AI-powered workflows (JSON, LLM decides navigation) and deterministic flows (YAML, fixed sequences, no LLM calls).

- ADB over WiFi + Tailscale for remote control. The phone becomes an always-on agent you can trigger from anywhere.

- Supports Groq (free tier), OpenAI, OpenRouter, Bedrock. Ollama support just landed for fully local inference.

- One-line install: curl -fsSL https://droidclaw.ai/install.sh | sh

Built with Bun + TypeScript. 35 example workflows included covering messaging, social, productivity, research, and lifestyle tasks.

https://github.com/unitedbyai/droidclaw https://droidclaw.ai

Comments URL: https://news.ycombinator.com/item?id=47037459

Points: 3

# Comments: 0

I built DunSocial because I kept ghosting my own social media. Had things to share but hated the process.

Every scheduler assumes you have content ready. I never did.

So I built one that writes for you. It learns how you talk. Remembers what you like, what you'd never say. Shows up with drafts. You tweak, schedule, done.

Built our own memory infra to make the voice learning work.

The memory part was the hard problem. Not just "tone: professional" dropdowns. Actually learning from how you write over time. Still early. Using it for my own stuff.

Would love feedback. dunsocial.com

Comments URL: https://news.ycombinator.com/item?id=46576268

Points: 1

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=46391658

Points: 1

# Comments: 1

That question invites vague feedback from people who don’t use your thing. You want behavior. Friction. Patterns.

Wrote a short note to remind myself too!!!

What's your thoughts on this?

Comments URL: https://news.ycombinator.com/item?id=43561559

Points: 2

# Comments: 1

Everyone talks about AI replacing coders and writers.

But what if the real shift is when it starts to replace taste?

Comments URL: https://news.ycombinator.com/item?id=43526710

Points: 2

# Comments: 1

- Sanju

Comments URL: https://news.ycombinator.com/item?id=42497718

Points: 4

# Comments: 1

+1 Agreed. Plans fail when sticking to them matters more than adapting...

TBH, the best plans know when to shift.

Flexibility > Rigidity

Success rarely comes from what we aim for—it’s often about timing & flow. Perfection can hold us back, but staying open leads to breakthroughs.

Few irl examples:

1. Spend hours perfecting a video, 100 views. Post a raw clip, and it goes viral.

2. A polished product flops, but a scrappy weekend tool takes off.

3. A masterpiece gets ignored, but a casual sketch goes viral.

4. A failed dish turns into the next food trend.

Don’t wait for perfect. Keep creating, stay open, and let the unexpected happen.

P.S. Success is personal. But if you aim for something and it doesn’t happen, that’s a failure in that context.

- What was invented by accident?https://www.reddit.com/r/AskReddit/comments/ygqwwl/what_was_...

- What is the best accidental invention?https://www.reddit.com/r/AskReddit/comments/1cajrnr/what_is_...

When you let things flow instead of sticking to a rigid plan, you end up finding better ideas. It’s like the pressure’s off, and the real work happens naturally. That freedom makes everything more productive ;)