ShaggAi is a local-first AI platform that turns natural language into real actions 🧠⚙️ powered by Python scripts and workflows on Windows and Linux. Create folders 📂, rename and clean up files 🧹, run commands and automations. It also adds ghost text autocomplete ✨ to many Windows apps/editors: type shagg: and let it finish your thoughts ⌨️.
No reviews yetBe the first to leave a review for ShaggAi - Turn Your Words Into Action
Maker
📌
Hey Product Hunt! 👋
I built ShaggAi because I was tired of doing the same boring things on my computer over and over, and copy-pasting between chatbots and my editor. I wanted one local tool that could both do things on my machine and help me write faster.
What ShaggAi does
1️⃣ Natural language → real actions (Python under the hood)
ShaggAi turns plain language into actions on your PC, backed by Python scripts and workflows on Windows and Linux.
Right now I mostly use it for things like:
Creating year/month/day folder structures
Renaming and sorting messy files
Cleaning up directories
Running small scripted tasks
It generates Python-based actions, runs them locally, and lets you see what happened so you stay in control.
👉 I don’t think I’ve even scratched the surface of what people might want here. I’d really love to know:
What kinds of actions would you turn into “one sentence → done”?
Dev workflows? System cleanup? Backups? Browser automation? Something else?
I really miss use cases for Speech to Action really... The most thinks you dont need daily sometimes its pretty handy but still... maybe you have more ideas for use cases 😅
2️⃣ Ghost text autocomplete in your editors
On Windows, ShaggAi overlays ghost text suggestions in many editors:
- Notepad++
- Outlook
- Word
- Many browsers / text areas
Type shagg: and let your editor finish your thoughts, :shagg to turn it off. You can:
- Accept a suggestion with a shortcut
- Accept the full completion
- Regenerate until it sounds right
It works for code, emails, docs, and notes without forcing you into a new app.
A few nerdy bits
🖥️ Local-first: actions run on your machine.
🧠 Bring-your-own models: use local / self-hosted LLMs if you want max privacy.
🧩 Custom system context: teach it your tone, rules, and personal details.
🔐 Privacy-aware filtering: it tries to filter passwords, API keys, etc. (for sensitive work I still recommend local models).
What I’d love feedback on
- Where ghost text feels most useful to you (emails, docs, code, notes?).
- Which natural-language → action use cases you’d actually you know or care about , is text to action even useful??
Thanks for checking it out 🙏
You can try ShaggAi here: sha.gg or shaggai.app