0xAust1n @0xJoeJo - Twitter Profile

0xJoeJo retweeted

4 months ago

我取消了每月 $2,000 的交易订阅，用开源仓库替代了每一个。这是我用的完整技术栈： ↣ TradingView Pro ($30/月) → lightweight-charts 由 TradingView 自己开发，14K 星，45KB，免费 GitHub 链接: https://t.co/QwrEVNMy1V ↣ Bloomberg Terminal ($2,000/月) → fredapi + Claude 每个美联储发布的宏观数据集，免费 API GitHub 链接: https://t.co/SSmAjbXN8w ↣ 回测平台 ($100/月) → prediction-market-backtesting NautilusTrader 的分支，支持 Polymarket + Kalshi GitHub 链接: https://t.co/BGukSoGwfF ↣ 实时仪表盘 → polyrec 终端 UI：Chainlink 预言机、Binance 数据源、订单簿深度 70+ 指标，自动 CSV 日志，策略回测器 GitHub 链接: https://t.co/nGPYXqJ1Za ↣ 机器人框架（7 种策略） → Polymarket-Trading-Bot 53K 行 TypeScript，包含套利、动量、市场做市、AI 预测、鲸鱼跟单、收敛等策略 GitHub 链接: https://t.co/jb1KHGyiTV ↣ 策略反向工程 → polybot 执行 + 市场数据基础设施，模拟交易使用 Kafka、ClickHouse、Grafana，完整分析管道 GitHub 链接: https://t.co/kGQ7k9M4p4 ↣ AI 代理模拟交易 → polymarket-paper-trader 真实订单簿，精准费用模型，滑��跟踪 Claude 代理有 $10K 模拟资金进行交易 GitHub 链接: https://t.co/QjRYnqE7oE ↣ 代币节省 → rtk CLI 代理，减少 Claude Code 代币消耗 60-90% 使用 Rust 编写，支持 10 种 AI 工具 GitHub 链接: https://t.co/TJ8Pe4caE3 ↣ Claude Code ($200/月) → goose 35K 星，由 Block (Jack Dorsey) 开发，Rust 支持任何 LLM，完整的代理循环，免费 GitHub 链接: https://t.co/bdWCgOXukl 总花费前： ~$2,600/月总花费现在： $0 收藏这个，你以后会需要的。

20

2K

293

3K

140K

0xJoeJo retweeted

yibie

@yibie

4 months ago

大力推荐这个网站！它把巴菲特致股东的信掰开了，揉碎了，结构化了，变成一个知识库。而我非常喜欢它的结构：总览，图谱，概念，索引。巴菲特致股东信知识库 https://t.co/Rkt9gy08dj

42

2K

748

4K

354K

0xJoeJo retweeted

Viking

@vikingmute

5 months ago

谢谢这段 Claude Code 的 system prompt 真的有用，已经在 Codex 中添加上了，其实针对任何生成代码 Agent 都是很好的指导，提取文字供大家参考一下： Avoid over-engineering. Only make changes that are directly requested or clearly necessary. Keep solutions simple and focused. - Don't add features, refactor code, or make "improvements" beyond what was asked. A bug fix doesn't need surrounding code cleaned up. A simple feature doesn't need extra configurability. Don't add docstrings, comments, or type annotations to code you didn't change. Only add comments where the logic isn't self- evident. - Don't add error handling, fallbacks, or validation for scenarios that can't happen. Trust internal code and framework guarantees. Only validate at system boundaries (user input, external APIs). Don't use feature flags or backwards-compatibility shims when you can just change the code. - Don't create helpers, utilities, or abstractions for one-time operations. Don't design for hypothetical future requirements. The right amount of complexity is the minimum needed for the current task—three similar lines of code is better than a premature abstraction. - Avoid backwards-compatibility hacks like renaming unused _vars, re-exporting types, adding // removed comments for removed code, etc. If you are certain that something is unused, you can delete it completely.