Hi Product Hunt, I'm Guofeng (misswell), a solo indie developer. Over the past few years I've shipped 16 small, focused tools. Each one solves a problem I actually had. No team, no VC, no growth hacks, just software I wanted to exist. Here's the full collection.

macOS

• MacPilot - native macOS toolbox: auto-quit apps, clipboard history, window switcher, screenshot OCR, BLE proximity unlock. Free and open source. github.com/misswell/MacPilot

• OctoShrink - local-first image compression. Drop in images or folders, batch compress, nothing uploaded. github.com/misswell/octo-shrink apps.apple.com/cn/app/id6792604654

• DevSweep - cleans developer caches to make room for your next build. Never touches source. github.com/misswell/DevSweep

• ClipNest - one shortcut sends clipboard content into your Obsidian vault, with AI sorting it into Markdown notes. github.com/misswell/ClipNest

• DeepSeek Harness Desk - an independent open-source desktop client for running DeepSeek Harness in a calm, focused window. github.com/misswell/deepseek-harness-desk

• Clawdesk - a quiet, visible desktop presence for your coding agents. github.com/misswell/Clawdesk

• CCRBar - control Claude Code Router from the menu bar. Pure SwiftUI, no Electron. github.com/misswell/CCRBar

iPhone and iPad

Hi PH, I m Andriana, co-founder of k dlo.

k dlo is a sales workspace for people who build B2B products.

After our launch here we looked at our data and saw a clear pattern: asking for a card was stopping people before they even saw what the product does. So we acted on it. k dlo now has a 7-day trial with no card.

This isn t an easy decision for a small team with no funding. Every trial costs us real money in AI tokens. We re doing it because right now one honest opinion from a real user is worth more to us than anything else!

A few days ago, I posted about how easy it is to get carried away building things, especially with AI making implementation so accessible.

It reminded me of something I went through with Atlas, my travel app project.

I spent quite some time trying to improve place search, exploring different search engines and data providers. But eventually I realized that getting anywhere close to perfect search would take a lot of effort and potentially be quite costly. Hard to justify when you don't even have users yet

So I took a simpler approach: let users click anywhere on the map and add a place themselves.

I m learning electronics and soldering, so I decided to challenge myself by building a simple, budget-friendly alternative to some Flipper Zero functions.

The project uses a D1 Mini ESP8266, an OLED display, IR receiver and transmitter modules, 433 MHz RF modules, an 18650 battery, HiveMQ, MQTT, and a private web dashboard.

Hello Product Hunt! 

I m Shake

Claude Code makes it easy to ship interfaces quickly, but builders are often too familiar with their own product to notice where a first-time user gets confused.

I m working on HallwayTest, an MCP workflow that helps Claude Code builders put a product flow in front of a real person and bring the usability findings back into the into your Claude Code, refine, then repeat again till UX polished.

I m looking for builders with a working web product with complex UX, who are willing to try the workflow and tell me where the setup or results are confusing.

You get:

Hi PH! I built Advance Poster, a Chrome extension that automatically posts to your Facebook groups.

Write one post, pick your groups, and it publishes everywhere with pauses between posts, a unique wording for each group, and a link to every post. It can also write texts and create images with AI.

Free for 5 days, no card needed. I'd love your honest feedback: what's missing for you?
Advance Poster: post to all your Facebook groups in one click (1-minute overview)
Get Advance Poster

Hi everyone, I'm Jim, an independent iPhone developer. Cognilo launches on Product Hunt tomorrow, Oct 21.

Five puzzles. About six minutes. Your next personal best.

- The Daily 5: five short brain games picked for your level across memory, focus, logic, numbers, spatial and language.

- Untimed by default; choose Precision, timed Sprint or unrated Zen.

Hi, I m Martin from RealTestData. I ve founded RealTestData s.r.o. last autumn and I m looking for feedback here. RealTestData is a B2B product to help companies with test data by generating synthetic test data offline in seconds. All details are on the web page. Any recommendations how to improve my LinkedIn presence?

Thanks and happy new year

Martin

I'm building MediDue, a planned English-language iPhone app for home medicine inventory. The idea starts with a practical question: could a simple list make it easier to check what is at home, where it is stored, how much is left, and which expiry dates need a closer look?

I currently have a browser prototype. It records medicine names, expiry dates, quantities and storage locations, with views for inventory and expiry or low-stock tasks. Users check the information they enter. The iPhone app is still in development, with no public download available yet.

The design problem I'm working through is the effort needed to keep those records useful. Entering a quantity once is straightforward; keeping it accurate as things get used or moved is harder. Adding more fields might make the first record more complete while also making every later update more work. I don't have user research results to settle that trade-off yet.

I'd like to hear from people who have kept a household inventory, or built a tool people use only occasionally: