josharsh
A minimal browser agent built on the accessibility tree
Browser agent that reads a page in ~2k tokens, not ~180k
Show HN: Ez FFmpeg – Video editing in plain English
I built a CLI tool that lets you do common video/audio operations without remembering ffmpeg syntax.Instead of: ffmpeg -i video.mp4 -vf "fps=15,scale=480:-1:flags=lanczos" -loop 0 output.gifYou write: ff convert video.mp4 to gifMore examples: ff compress video.mp4 to 10mb ff trim video.mp4 from 0:30 to 1:00 ff extract audio from video.mp4 ff resize video.mp4 to 720p ff speed up...
Show HN: AI Terminal Course – Learn AI Engineering Like Learning Vim
https://www.npmjs.com/package/ai-terminal-course
I spent months watching AI tutorials on YouTube. Took notes. Felt like I learned a ton. Then I sat down to build something and... couldn't. I didn't actually understand a lot of stuff!The problem: Passive learning doesn't work for engineering skills.What makes this different:• You TYPE every command (no copy-paste) • You SEE real API responses (not fake demos) • You PAY real costs (~$10...
Show HN: Md-pdf-md – Bidirectional Markdown<>PDF with local AI vision
https://www.npmjs.com/package/md-pdf-md
Convert Markdown to beautiful PDFs AND extract Markdown from PDFs using local AI. Features 4 themes, VS Code syntax highlighting, and LLaVA vision model. Zero config, privacy-first, completely open source. No cloud, no API costs.