You can now turn any technical book or document into a Claude Code skill 🤯
Dumping the raw PDF into Claude costs ~200K tokens before you ask a single question. book-to-skill compiles the book once, then loads only the chapter you need.
→ Per-chapter summaries pulled from the source
→ Glossary + cheatsheet of named frameworks
→ Docling keeps tables, formulas, and code as markdown
100% Open Source.
Boris Cherny, the creator of Claude Code at Anthropic, just explained how he runs thousands of agents from his phone while he sleeps
in this talk he breaks down exactly how he manages thousands of agents from his phone:
- the 14% you lose to CLAUDE.md before typing a word
- the architecture behind running a few thousand agents overnight
- why the system matters more than the tool
- the knowledge structure that makes everything findable, connected, and useful
if you've been using Claude for months and every conversation still starts from scratch with zero memory and zero context, you don't have a system. you have a chat history
instead of another show tonight, watch this
make sure to bookmark it before it gets lost in your feed
the guide is in the article below
1/5
I'm a cardiologist. I have spent twenty years watching cholesterol destroy arteries, trigger heart attacks, and kill people I care about.
Today, Eli Lilly presented data that may begin to end that era.
VERVE-102. A single infusion. One dose. It uses base editing to permanently turn off the PCSK9 gene in your liver.
Presented today at the European Atherosclerosis Society Congress:
88% reduction in PCSK9.
62% reduction in LDL cholesterol.
Sustained up to 18 months.
No treatment-related serious adverse events.
One infusion. Not daily pills you forget to take. Not monthly injections. One dose — and your cholesterol may stay low for the rest of your life.
I've been trading since 2006, and I don't want to leave this earth without handing down most of what I've learned — to my kids first, and to anyone willing to learn.
They're too young to understand it now. So the best way to preserve it is to bring what my eyes see onto the chart itself.
In other words: build indicators that make my setups easy to spot — clean pivots, hidden/quick pivots, advanced setups, staying off the left side of the base, shakeout & reclaim, and more. Same spirit as what I've already done with Volume Pocket Pivots and Moving Average Trend.
It's a big project. But I played with the concept this morning, and I think I can pull it off.
All the hard work I've put in over the years will not go gently into that good night.
I'll make videos of the progress to demonstrate how i am bringing the entire strategy together. I'll figure out how to make it available to the community at some point in the future — I just want it to outlive me and actually help people.
This won't be your typical "indicators". What I want to create is multiple tools that when you layer them on the chart, it helps you "see" what I see.
I have 4 fake teeth,
2 root canals,
and more than 12 teeth with fillings!
For 20 years, I dreaded every dentist visit.
Then a machine I ordered from the US
became a ₹1Cr business in 2 years.
This is the most personal story I'll ever tell:
#ClaudeCode#AlgoTrading#PineScript#NIFTY#MCP
1/ I connected TradingView to Claude Code via an MCP server on my Mac. It now reads my live charts, writes Pine Script, and iterates on strategies — all from the terminal. Full setup + why it matters 👇
2/ The problem this solves:
For months I was screenshotting charts into AI chats, pasting Pine Script back and forth, losing context between conversations. Slow, error-prone, and the AI never saw what I was actually looking at.
MCP fixes that. Claude becomes a live participant in your TradingView session.
3/ Setup (one-time, ~10 mins on Mac):
• Launch Chrome with CDP debug mode (--remote-debugging-port=9222)
• Open TradingView in that Chrome instance and log in
• Drop a project-level .mcp.json in your working dir pointing to the TV MCP server
• Start Claude Code in that directory
• Run tv_health_check — confirms Claude can see the active chart
4/ What the MCP server actually exposes:
• Current symbol, timeframe, and indicators on the chart
• Pine Script editor read/write access
• Chart annotations, drawings, alerts
• The ability to switch symbols and timeframes programmatically
Claude isn't guessing what's on your screen — it's reading it directly.
5/ How I'm using it for intraday on NIFTY:
• "Read the 15-min ORB on this chart and flag if VWAP is supporting the breakout" — Claude pulls the levels and gives a structured read
• "Refactor this Pine Script to add a 1% per-trade risk and 3% daily loss cap" — edits the script in place
• "Backtest this on the last 60 sessions and summarise win rate by hour" — runs and reports
6/ Pine Script dev workflow is the biggest win:
Claude writes v1 → I paste into TV → it sees the compile error or the backtest output → fixes it → loop.
No more copy-pasting tracebacks. Iteration time on a new strategy went from days to hours.
7/ Strategies I've built or refined this way:
• 15-min ORB + VWAP with partial exits and position sizing
• Evening screener with IBD-style RS ranking
• Hard-coded intraday risk rails (1%/trade, 3% daily, 14:30 IST exit)
All iterated conversationally against live charts.
8/ Caveats worth knowing:
• CDP mode means that Chrome instance is exposed locally — don't run it on shared networks
• The MCP server is community-built, not official TradingView — expect rough edges
• Heavy chart manipulation can lag; keep the chart simple while Claude is reading it
9/ Why I think this matters beyond my own setup:
The gap between "AI that talks about your charts" and "AI that works with your charts" is huge. MCP closes it. Same pattern will hit Bloomberg, Refinitiv, broker terminals — anywhere a trader spends their day.
10/ If you do any serious Pine Script work or strategy dev on TradingView, spend the 30 mins.
Using TV+CC+MCP combination, You can do things now in less than 10 minutes , what used to be a 1 year activity in the past …..
Bookmark this anti-FOMO manifesto that I'm giving you for free & read it every morning till the last trading day of your life.
⦿ I have made peace with the fact that I have neither the capacity nor the need to capture every upmove in the market.
⦿ I like to trade only proper constructive setups, & not enter here-&-there in the name of what kind of structures are trending currently.
⦿ If the stock goes up without giving me a tradable setup, let it go; it was never mine in the first place.
⦿ I will allow myself to feel FOMO only when I actually miss a setup that met my tradability criteria, either due to a deficiency in my scanning or in my execution. The result is immaterial. Even if the stock had later hit my stoploss, and it was a setup worth trading (under appropriate market conditions), it was a miss on my part.
⦿ I don't care what this world thinks about me, because I don't think anything about it. I judge myself not by the quantum of money I earned, but whether my approach was process-driven or not.
Rotation people!! How to spot it:
Think back to january...we witnessed a HUGE rotation from software into semiconductors.
Thinking logically...that big $$$ that rode semis to ATHs is going to eventually take profit and rotate elsewhere.
Not saying the semiconductor run is over, but software seeing some renewed strength makes me think we're in the early stages of a short term rotation out of semis.
The question is...into where?
First, how can you tell when rotation is happening?
- former leaders stop reacting well to good news
- relative strength starts shifting
- watching major ETF flows (see screenshot)
- leadership narrows before rotation (1-2 stocks carrying the whole theme)
- new narrative emerges + sexier story
- expansion happening in a different theme
The caveat is don't be too early to rotate out of the strong theme...wait for confirmation of weakness + lower highs.
Some sauce:
> go to finviz
> make a watchlist of all the top ETFs
> rank by performance (week, month, Q)
> see where the strength has been on longer timeframes
> see where the strength is starting to appear on shorter timeframes
Gives you an idea on where $$$ has been + where $$$ could be rotating into...so you can focus some attention to potentially find the next leading sector.
The BIGGEST CAPEX Cycle in Human History is LOADING...
And if you’re not participating - you're missing on the ALPHA of a lifetime
2 Indian Small Caps, positioned in the centre of this MEGA CYCLE, are quietly hitting all-time highs, surging over 100% in the last 6 months
This is only the BEGINNING…..
A thread
🧵🧵🧵
#StockMarket #investing
What if I told you that some of my best trades didn't have a setup?
“No setup, no trade” is great advice...but for me its a little different: If I can’t define my risk, I don’t trade. Period.
The market isn’t always clean. Patterns don’t always line up perfectly. But here’s the rule that never changes: I know exactly where I’m wrong before I hit buy. If I can’t define that level? The trade doesn’t happen. End of story.
That risk level can be anything... a support line, a prior high, even a psychological price point. It doesn’t have to be fancy, but it must exist. If I don’t have a clear exit if I’m wrong, I’m not trading. I’m gambling. And I don’t gamble.
The bottom line is setups are secondary. The real edge in trading is risk management. The question isn’t “Is this a setup?” It’s “Can I manage risk here?”
If the answer is no, you walk away. The market will always give you another shot.
Discipline wins. Always. Define your risk, or don’t trade.
Came across this document of Qullamagi’s 2020–2021 trade positions.
If you go through these trades, one of the best things you’ll notice is the risk–reward. There a lot of losing trades. But the losses are not big, mostly around one or two X.
On the other hand, the winning trades are much bigger.
If you go through the stock names, try to understand the thought process behind each trade.
Yes, this is from 2021, but a lot of things can still be understood today if you go through these trades.
Go through it. Try to understand it. Good luck.
Link - https://t.co/7Yl0qzb1AW
Dr. James pennebaker spent 30 years studying why some people recover from trauma and others don't.
He found the same pattern in every person WHO healed.
It had nothing to do with therapy or time.
Medicine ignored it. His patients didn't. 🪡
Qullamaggie on The Smaller Your Trading Account, The More You Should Look for Smaller Moves and Compound
“If I had a smaller account, I would be looking for smaller moves and making more trades and compounding that way. Like at this point, I’m a home run trader. I don’t care for pennies. I want the big trades, and they don’t come often—they don’t come around too often. So if you have a smaller account, you need to be more aggressive.”
Robert Sapolsky is a Stanford neuroscientist who proved chronic stress is the silent killer doctors ignore.
On Chris Williamson's podcast, he revealed 10 "normal" habits you do every day that wreck your sleep, mood, and nervous system:
1) Replay conversations in your head
This 2 hour lecture by Yann LeCun (Turing Award winner) will teach you why the next trillion dollar AI company won't be built on LLMs.
He trashes the $100 Billion LLM race, attacks Musk and Amodei, declares scaling dead.
Bookmark & watch tonight after work, skip to 7:00.
For over six decades, India's passenger trains ran on the same basic idea. A powerful locomotive up front, pulling a long string of passive coaches behind it. As of March 2018, 49,033 conventional coaches were still in service. Then, in January 2018, the ICF flagged off its last one.🧵👇
this Columbia Business School lecture on price momentum will change how you think about trading.
he explains the exact portfolio construction behind big hedge funds, why momentum is so powerful, and a strategy that turned $1 into $41,000.
25 minutes. free. bookmark it.