Pumpusha

[Nico @nicos_ai]

DEJA DE PROMPTEAR. EMPIEZA A LOOPEAR.

Encontré un sitio que recopila los loops más usados por la comunidad

→ https://t.co/jNdeuXX9SW

Los propios creadores de Claude Code lo dicen: el futuro no es promptear, es diseñar loops.

No sabes qué son ni cómo crearlos?

Este articulo lo explica detalladamente👇

[Kotte @Gustafssonkotte]

Polymarket cheat code update

I went through every trade my machine refused to take and asked one question: was that money left on the table, or was the refusal correct?

THE AUDIT:

- 58 windows stopped by the trend-guard. Raw win rate of those signals: 90%. Sounds like stolen profit, until you look at the prices. Almost all of them cost 0.87-0.98, a zone where even a 90% win rate earns nothing. In the buyable zone (0.85 and below) those same blocked signals win only 56% of the time. The guard cuts exactly the junk and nothing else. The math confirmed its right to exist.

- 19 cut by the price ceiling. Average cost 0.94, win rate 94%. Edge = zero, accurate to the cent. That number deserves a pause: the market reprices momentum instantly and almost perfectly. There is no free money in the obvious moves.

- 4 signals reached the order at 0.78-0.85. That is the whole game. Two or three buyable signals per day, each alive for seconds before the market closes the door.

THE UPGRADE OF THE DAY:

A momentum signal and an overbought signal are the same event seen from two sides. RSI at 70 after an $83/min burst is not a warning of reversal. It IS the burst. You cannot have a momentum entry without overbought conditions, by definition.

The signal itself is validated at 88-90% across 162 windows, and that validation already includes every overbought tick. Which means filtering bursts for overbought conditions cuts edge the data has already paid for.

So I split them apart:
- Burst entries now go straight through.
- Slow drift stays guarded. On drift that same filter caught a snap-back this very morning and kept the money.

Every filter now works exactly where it proved its worth. And only there.

ONE MORE THING THE LOGS REVEALED:

Fills come in cheaper than the screen. Screen shows 0.83, the fill lands at 0.75. That is 8 extra points of edge no backtest will ever show you, because backtests trade at screen prices. Only a live machine sees the real cost of entry.

Win rate by itself means nothing. Your entry price IS the win rate you must beat. Pay 0.82 and guess right 82% of the time, you have earned exactly zero.

The market does not hide the edge. It prices it. Your job is the few seconds when it has not priced it yet.

[Suryansh Tiwari @Suryanshti777]

Claude Code feels completely different once you install this.

Anthropic quietly released an official plugin called claude-code-setup and it basically turns Claude Code from “pretty good” into an actual AI dev environment.

It scans your project and recommends:

→ hooks
→ skills
→ MCP servers
→ subagents
→ automations

Then sets everything up step-by-step for you.

Most people are using Claude Code completely vanilla…

which is why their experience feels messy.

The real power comes from the ecosystem around it.

Install:

/plugin install claude-code-setup@claude-plugins-official

Bookmark this before you forget it.

[Himanshu Kumar @codewithimanshu]

Claude is now controlling TradingView live from my terminal.

Switching symbols. Writing Pine Script. Batch scanning futures. Replay trading. Drawing levels.

All autonomous. Zero clicks.

Still has rough edges but the vision is crystal clear.

I told it:

Find me every BTC futures contract with RSI below 30 and volume spike above 200%.

14 seconds later:

→ 6 contracts identified
→ Charts loaded
→ Support levels drawn
→ Pine Script backtests running
→ Entry zones marked

Didn't touch the mouse once.

Then I said:

Replay last week. Show me where your system would have entered.

It switched to replay mode. Scrolled through price action. Marked every edge. Calculated P&L in real-time.

$4,780 theoretical profit from 9 trades.

83% win rate.

Now it writes custom Pine indicators on command:

Build me a momentum oscillator that tracks whale wallet activity correlated with price.

40 seconds. Script deployed. Indicator live on chart.

Most traders are still clicking through 50 charts manually.

Claude scans 200+ in under a minute.

Finds the setups. Draws the levels. Backtests the edge. All while you watch.

This is not about replacing your strategy.

It's about executing it 100x faster.

You only need Claude + laptop + 1 hour/day.

Giving This Free for 24 hours. To get it:

1. Comment the word CLAUDE
2. Like and Retweet this post
3. Follow me @codewithimanshu (so i can DM you)

Save this post. Deploy this setup this weekend. Start testing. Scale on evidence.

[0xFuck.ETH @0x404page]

卧槽兄弟们，GitHub 上真的有一堆免费到离谱的项目。

很多能力已经能直接干掉你正在月付的软件。

1. TradingAgents
AI 多 Agent 量化交易框架
https://t.co/TUb3q0uILN…

2. LibreChat
一个界面接入 ChatGPT、Claude、Gemini 等多个模型
https://t.co/UUMftLSOU4…

3. HyperFrames
HeyGen 开源的视频生成引擎
https://t.co/OZKS8R1qnj…

4. Fincept Terminal
开源版彭博终端
https://t.co/3QGYCnWJqV…

5. MoneyPrinterTurbo
AI 一键生成短视频
https://t.co/XenuwHKMwg…

6. Agentic Inbox
Cloudflare 开源 AI 邮件助手
https://t.co/g92N5IMxIG…

7. VoxCPM
AI 声音克隆工具
https://t.co/nWDB88AF0m

8. Flowsint
开源 OSINT 情报分析工具
https://t.co/dNp65L5v0e…

9. agent-skills
Claude Code 技能库
https://t.co/y7RnzfKJT8…

10. Nango
开源 API 集成平台
https://t.co/xBK94t6Pgj

兄弟们，这些真不是玩具项目。

很多你还在交月费的软件，GitHub 上已经有人做出开源平替了。

一句话：

别只会收藏 AI 工具网站了。

真正狠的东西，很多都藏在 GitHub 里。

[noah lewis @novalevys]

@theonejvo

[Jamieson O'Reilly @theonejvo]

Same intelligence, different artifact. The prompt alone moved branding, voice, section structure, and overall feel.

[Jamieson O'Reilly @theonejvo]

Same prompt to each: "create a modern Apple style landing page."

[Jamieson O'Reilly @theonejvo]

I dropped it into Claude Code using claude --dangerously-skip-permissions --system-prompt-file CLAUDE-FABLE-5.md and ran plain Opus 4.8 in the other pane as a control. Same model in both panes (both show "Opus 4.8 · 1M context"), so this is purely a system-prompt comparison.

[Jamieson O'Reilly @theonejvo]

Pliny posted what's claimed to be the Fable 5 system prompt.

[Jamieson O'Reilly @theonejvo]

Did I just unlock claude-fable-5-lite? 😂

Since Fable 5 got pulled (US export control order, Anthropic is contesting it), I wanted to see how much of its character lives in the system prompt vs. the model itself.

I ran the leaked Fable 5 prompt on Opus 4.8 head-to-head against stock 4.8.

[Punisher @0x_Punisher]

It's been 2 months since I launched my sweeper bot.

Since then, i claim $500 EVERY SINGLE DAY.

I dropped a full guide so everyone can build the same one.

Thousands scrolled past.

Those who read full now print passively.

The choice is yours.

[Blockchain Surfer ☝️ @0xSurferX]

"Trade Polymarket while you sleep" is FAKE.

No bot can work 24/7 without human watching him.

It will trade your money away while you sleep.

I count four things in this file that drain wallets.

Look at the code:

> It runs every 5 minutes on a cron

Read the line under the code.

A timer wakes up, checks the market, maybe trades.

On Polymarket that is geologic.

By the time the cron fires and the model finishes thinking, the move already happened and the fill is gone.

This has to listen to a live feed, never wake on a clock.

> It asks an AI to predict a coin flip

The system line literally says "you are a prediction market analyst"

The task says "estimate real probability" that BTC goes up.

On a 5-minute window there is nothing to predict.

It's noise.

You're paying an API to guess a coin toss, and guess it differently every time you ask.

> The model places the order itself

Look at the last two lines.

If confidence is over 0.7, the LLM pulls the trigger.

A confidence score of 0.88 is not an edge.

It's a vibe with a decimal point.

You just handed the trade button to the least deterministic thing in your whole stack.

> They told you the prompt is your strategy

The article calls this file "where your strategy lives"
It isn't.

A prompt asking AI to guess a probability is not an edge.

A real edge is reproducible, testable, and structural - latency, queue position, reading a number that's already decided.

None of that lives in a system prompt.

So what does the real version of these same 20 lines do?

It listens to a live feed instead of waking on a timer.

It reads a decided fact instead of predicting a coin flip.

It fires a pre-signed order on a fixed rule, not on a model's mood.

And it keeps the AI on plumbing, building and parsing, never on the bet itself.

Same length.
Completely different bot.

Simple is good.
Simple and wrong just loses faster.

Leaving my full guide on building this the right way quoted below.

Eery mistake in this file is covered in it.

Good luck.

[Willy Drucker @WillyDrucker]

@OpenRouter I believe the results!

Built an extension for VS Code that allows Claude and Codex to iterate back and forth and the results are mind blowing.

[OpenRouter @OpenRouter]

Fusion is neurodiversity, but for models. Try it now!

💬 Chatroom: https://t.co/8Fyf0tjGa5 (pick a preset or build a custom panel)

⚙️ API: docs at https://t.co/170CR9bu07

ℹ️ More info on the blog post: https://t.co/jjB95vm7yY

[OpenRouter @OpenRouter]

Introducing the Fusion API, the smartest compound model in the market.

Fusion achieves Fable-level intelligence at half the price.

How it works 👇

[simulx4 @simulx4]

@ClaudeDevs

[Jack Phantom @0xJackPhantom]

In this video, an AI trading agent on Liteverse uses paper trading mode. The agent analyzed the market, opened the trade, managed the setup, and reached profit in under 3 minutes.

This is not real-money trading and not financial advice. It’s a risk-free paper trading demo showing how autonomous AI agents can be used for market analysis, trade planning, and simulated execution.

[Jack Phantom @0xJackPhantom]

@claudeai Yes, I completed my app Liteverse in 2 days with Fable 5. This thing is different. It doesn’t feel like an assistant anymore, it feels like a real builder. 😅

> https://t.co/eTyfXuNPig

[Claude @claudeai]

https://t.co/48ZXHCOSiJ