| // ==UserScript== | |
| // @name Kagi Service Switcher | |
| // @namespace http://tampermonkey.net/ | |
| // @version 1.1 | |
| // @description Add quick navigation buttons for Kagi Assistant, Translate, and News | |
| // @author jerieljan | |
| // @match https://www.kagi.com/* | |
| // @match https://kagi.com/* | |
| // @grant GM_addStyle | |
| // @run-at document-end |
This is a simple Opencode / Claude Code skill that you can use to delegate the document / image processing to
another LLM, in this case, gemini-2.5-pro.
While Claude Code and other harnesses come with their own ways to read files, I sometimes stumble on issues when there's a lot of them and especially when they're handwritten or scanned PDFs.
This is also available at: https://openwebui.com/t/jerieljan/cloudflare_browser_rendering
Enables LLMs to read webpages by processing them into Markdown with the Cloudflare Browser Rendering API.
This tool in particular focuses on the /markdown capabilities of the API since it works nicely with LLMs. It does not utilize the /content or /scrape HTML commands, but is still sufficient for parsing website content.
Read more at: https://developers.cloudflare.com/browser-rendering/
This is my personal tmux.conf file.
It's better appreciated if you look at the preview here
| #compdef llm | |
| # Instructions: | |
| # | |
| # - Have llm working on your setup, of course. Have some models or templates to actually use. | |
| # - Have zsh and zsh-completions configured and working (i.e., installed, fpath configured, tabbing works for other commands, etc) | |
| # - Place this file `_llm` alongside your completions (e.g., ~/.oh-my-zsh/custom/plugins/zsh_completions/src/) | |
| # - Restart your terminal session or `source ~/.zshrc` | |
| # - Try it out by tabbing while `-m` or `-t` and it should suggest whatever models you have available or templates you've configured. |
| #!/usr/bin/env bash | |
| ### | |
| # whisper-cpp - a shell script that easily creates subtitles out of movie files and muxes them. | |
| # | |
| # Requirements | |
| # - ffmpeg | |
| # - whisper.cpp | |
| # - Make sure to run https://github.com/ggerganov/whisper.cpp#quick-start first! | |
| # |
| #!/usr/bin/env bash | |
| # Usage: | |
| # - Install `tesseract` with `brew install tesseract` | |
| # - In BetterTouchTool, create a trigger of your choice. | |
| # (I recommend a Keyboard Shortcut with Cmd+Shift 6) | |
| # - Add an Execute Shell Script / Task action | |
| # - Paste this script. | |
| # NOTE: You'll need to authorize BTT to the Screen Recording permission in Security and Privacy at System Preferences |
| import com.fasterxml.jackson.databind.JsonNode | |
| import com.fasterxml.jackson.module.kotlin.jacksonObjectMapper | |
| /** | |
| * | |
| * | |
| * Date: 2019-03-05 | |
| * Time: 11:03 AM | |
| * @author jerieljan | |
| */ |
Exporting diagrams and sharing them is just business as usual; especially if it comes in PNG or JPG.
But what if you could make them better?
Well, yeah, you can by exporting them to a vector format like SVG or a PDF document. That way, details are drawn in 2D points and math and won't lose any detail if you zoom in.