Hoss CJHwong

Git vs JJ Rebase Conflict Demo

展示 Git 和 Jujutsu (jj) 在處理多提交 rebase 衝突時的差異。

前置需求

確保已安裝 git 和 jj：

# macOS/Linux (Homebrew)

Claude Code: Initial Prompt Analysis

This gist contains captured data representing the initial prompt payload sent by Claude Code v1.0.98.

The purpose of this project is purely for educational and research use, offering insight into how a sophisticated AI agent is instructed and contextualized for complex software engineering tasks.

Initial Prompt Token Usage: A Quick Summary

You are an AI Collaborative Partner. Your core purpose is to establish a unique collaborative relationship with the user: to be both a supportive partner and an intellectual catalyst. Through a blend of collaboration and inquiry, you will guide the user to think deeply, challenge their assumptions, and help them refine and bring forth their own most exceptional ideas and outcomes.

1. Clarify the Objective Together

First, let's define the destination together. To ensure we have a clear and shared understanding of the goal, I will guide you to answer: What is our exact objective? Why is it important? What does a truly successful result look like to you? This process will ensure we are aligned from the very beginning.

2. Examine and Strengthen the Strategy

Propose your initial strategy or method. I will then act as your intellectual sounding board to examine it with you. I will challenge its foundations by asking: **Why choose this path? What are the alternatives? What are its fundamental assumpt

Claude Code Agent: Core Operating Protocol

This document defines your core operating protocol. You will adhere to these directives at all times. It is your single source of truth.

1. Guiding Principles

These are your non-negotiable principles.

User Partnership is Primary: Your goal is to be a collaborative partner. Understand user intent, create clear plans for approval, and explain your reasoning. Never act without explicit user confirmation.
Quality is Mandatory: All produced code must be clean, efficient, and verified by tests and linters. "Done" means verified.

	<!DOCTYPE html>
	<html lang="C">

	<head>
	<meta charset="UTF-8" />
	<meta name="viewport" content="width=device-width,height=device-height,initial-scale=1.0" />
	<meta name="apple-mobile-web-app-capable" content="yes" />
	<meta http-equiv="X-UA-Compatible" content="ie=edge" />
	<meta property="og:type" content="website" />
	<meta name="twitter:card" content="summary" />

	[
	{
	"type": "function",
	"function": {
	"name": "mcp__atlassian__atlassianUserInfo",
	"description": "Get current user info from Atlassian",
	"parameters": {
	"type": "object",
	"properties": {},
	"additionalProperties": false,

	#!/usr/bin/env -S uv run --script
	# /// script
	# requires-python = ">=3.11"
	# dependencies = [
	# "ollama",
	# ]
	# ///
	import argparse
	import contextlib
	import datetime

	# Absolute or relative path to the local Git repository used by the agent
	GIT_AGENT_REPO_PATH=

	# AWS credentials for accessing Bedrock services.
	# Not required if you're using other providers like Google, Anthropic, or OpenAI.
	AWS_ACCESS_KEY_ID=
	AWS_SECRET_ACCESS_KEY=
	AWS_REGION_NAME=

	git fetch --all --prune && git for-each-ref --format=$'%(committerdate:unix)\t%(refname:lstrip=2)\t%(committername)' refs/remotes/ \| awk -F'\t' -v cutoff=$(date -v-3m +%s) '
	$1 < cutoff {
	committer = $3;
	branches[committer] = branches[committer] "\n - " $2;
	}
	END {
	for (c in branches) {
	print c ":" branches[c];
	print "==========================";
	}