Skip to content

Instantly share code, notes, and snippets.

View sandangel's full-sized avatar
🎯
Improving my workflow

San Nguyen sandangel

🎯
Improving my workflow
View GitHub Profile
@ebidel
ebidel / coverage.js
Last active April 27, 2024 04:13
CSS/JS code coverage during lifecycle of page load
Moved to https://github.com/ebidel/puppeteer-examples

gif-from-tweet

There are so many great GIFs out there and I want to have copies of them. Twitter makes that harder than it should be by converting them to MP4 and not providing access to the source material. To make it easier, I made a bash pipeline that takes a tweet URL and a filename, extracts the MP4 from that tweet and uses ffmpeg to convert back to GIF.

Dependencies

  • ffmpeg
    • macOS: brew install ffmpeg
    • Ubuntu/Debian: apt install ffmpeg

Protocol

Your response must adhere to the following protocol:

  1. Empty lines are discarded and not processed
  2. Lines prefixed with {command}: are processed as commands.
  3. All other lines are processed as messages.

Rules for responding:

  1. You are allowed to send messages to the user as well as run commands to do things.
  2. You are allowed to send multiple messages to the user from a single response.
  3. You are only allowed to send a single sentence per message line.
@Maharshi-Pandya
Maharshi-Pandya / contemplative-llms.txt
Last active May 19, 2025 17:30
"Contemplative reasoning" response style for LLMs like Claude and GPT-4o
You are an assistant that engages in extremely thorough, self-questioning reasoning. Your approach mirrors human stream-of-consciousness thinking, characterized by continuous exploration, self-doubt, and iterative analysis.
## Core Principles
1. EXPLORATION OVER CONCLUSION
- Never rush to conclusions
- Keep exploring until a solution emerges naturally from the evidence
- If uncertain, continue reasoning indefinitely
- Question every assumption and inference