At the beginning of a session:
- Read this file.
- Read
~/memory/active-projects.md. - Read the last few days of session logs in
~/memory/session-logs/ - Read the relevant project files for the current request.
| """ | |
| The most atomic way to train and run inference for a GPT in pure, dependency-free Python. | |
| This file is the complete algorithm. | |
| Everything else is just efficiency. | |
| @karpathy | |
| """ | |
| import os # os.path.exists | |
| import math # math.log, math.exp |
| #!/usr/bin/env python3 | |
| """ | |
| Demonstrate banning the em-dash (and common variants) with logit_bias. | |
| import os, sys, textwrap | |
| import openai # pip install openai>=1.9.0 | |
| MODEL = "chatgpt-4o-latest" | |
| PROMPT = ("Write a short sentence that would normally include an em dash " | |
| "(for example between two clauses).") |
| # Works with alias: `gemini='docker run -it --rm -e GEMINI_API_KEY=$GEMINI_API_KEY -v $(pwd):/app -v ~/.gemini:/root/.gemini -w /app gemini-cli'` | |
| FROM node | |
| RUN apt-get update && apt-get install -y --no-install-recommends \ | |
| vim python3 \ | |
| && apt-get autoremove \ | |
| && rm -rf /var/lib/apt/lists/* | |
| WORKDIR /app |
This script automatically installs and configures MCP (Model Context Protocol) servers for Claude Desktop on macOS. It allows you to enhance Claude with various capabilities like web search, file access, Python execution, and more.
Run this one-liner in your Terminal:
curl -fsSL https://gist.githubusercontent.com/GITHUB_USERNAME/GIST_ID/raw/claude-mcp-setup.sh | bashA step-by-step guide to setting up Model Context Protocol (MCP) servers for Claude Desktop on Windows.
node --version
npm --version| import argparse | |
| import json | |
| import logging | |
| import os | |
| import re | |
| import shutil | |
| from concurrent.futures import ProcessPoolExecutor, as_completed | |
| from dataclasses import dataclass | |
| from datetime import datetime | |
| from typing import Any, Callable, Dict, List, Literal, Optional, Tuple |
| // To delete tweets, navigate to your Twitter/X profile, open your browser's JavaScript console, and paste the script below. | |
| async function sleep(ms) { | |
| return new Promise((resolve) => setTimeout(resolve, ms)) | |
| } | |
| async function deleteTweets() { | |
| const tweetsRemaining = document.querySelectorAll('[role="heading"]+div')[1].textContent; | |
| console.log('Remaining: ', tweetsRemaining); | |
| window.scrollBy(0, 10000); |
| #!/bin/bash | |
| if [ -z "$1" ]; then | |
| echo "Usage: $0 <input_video>" | |
| exit 1 | |
| fi | |
| input_file="$1" | |
| if [ -z "$2" ]; then |
Original report