There are thousands of claude skills on github now.. and that's the problem..
because installing one is easy but finding the right one is the hard part..
you keep opening readme after readme, not even sure it fits what you're doing..
find-skills fixes that.
it's basically a skill that finds suitable and important skills for you.
its #1 on https://t.co/SY2YmUu1Co, 1.9M installs, made by the vercel team..
once it's in, you don't search anything.
you just tell claude what you need:
like
> is there a skill for better react? > find me one for pr reviews > i need to make a changelog
it looks through everything, finds the best match, and gives you the command to install it. one prompt, done..
best thing about it is that it only shows you the good ones. no random skill with 5 installs…
it checks how many people use it and who made it first. so you're not adding junk to your setup..
Setup takes about 1 min:
one command:
npx skills add vercel-labs/skills
that's it. now claude can find and install any skill for you while you work..
how i use it is:
1) starting something new "claude, is there already a skill for this?" it checks before you build from scratch.
2) want to search yourself
npx skills find <thing>
like npx skills find playwright.. it lists the best matches, you pick.
3) found one you like
"install it" claude adds it for you, done.
and yeah it works in all claude code, codex, cursor and 70+ agents.
Binance gave WLFI's USD1 stablecoin treatment most stables wait years for.
3 things happened in the same week:
1. BTCUSD1 perps live on Binance Futures, 100x leverage. first perp ever on Binance to use USD1 as the trading + settlement unit.
2. USD1 got the highest collateral tier on portfolio margin → 99.99%. same level as USDT and USDC.
that's not a casual tier assignment, that's Binance putting their reputation on USD1's peg.
3. USD1 now functions as quote currency, margin asset, AND collateral. full trading stack on a tier-1 venue. most stablecoins never get all three.
so basically, Binance placed USD1 in the same trust tier as USDT and USDC.
and it's all happening right after WLFI passed governance to burn 4.52B tokens and lock 62B more on a real vesting schedule.
catalysts are stacking quietly.
Higgsfield Supercomputer just got plugged into Gemini and the jump is wild.
a year ago AI video was basically a slot machine.
you typed, prayed, got something usable 1 in 5 tries.
text inside the video looked like a 4-year-old's handwriting.
now look at what this thing does:
> actually readable text generation inside the video
> motion graphics that look like a real ad agency made them
> frame-level editing so you can fix one bad frame, not regenerate the whole clip
> real-world search built in, so it knows actual facts and references
video attached.
this is what AI video is supposed to feel like → not a slot machine, an actual creative tool you steer.
Gonna be an important week for $BTC
- Testing the 100W EMA
- Testing the 200D EMA
At the same time the entire week is packed with events:
• US CPI (May 13)
• US PPI (May 14)
• Retail Sales (May 15)
• Trump-Xi / China headlines (May 15-16)
imo one soft inflation print or positive China headlines and we rip through those levels..
OpenAi just positioned GPT 5.5 directly at Claude Code's throat..
ChatGPT 5.5 is live with the exact language claude code uses.. "built for agents, uses tools, checks its own work, carries tasks through to completion"
GPT-5.5 lives in both chatgpt + codex.
Codex now runs on a model that's publicly benchmarked higher than opus 4.7 on agent tasks..
what this means if you build with AI:
> if you're on claude code max at $200/mo, test codex this week. even if you stay on claude, you need to know what's across the fence
> if you're building agents, gpt-5.5 + codex is suddenly the competitive stack to beat
> if you run MCPs, most work with both. swap providers in your config and benchmark side by side
the real winner isn't openai or anthropic. it's you, if you stop treating model loyalty like a religion and start routing tasks to whichever model wins that week.
> Close figma. Close canva. Cancel the adobe sub.
> open https://t.co/ZFDIq3FBuL
> Type the visual. ship the visual
> prototypes. slide decks. one-pagers. infographics. all from a chat box
> drop your brand in a Design System once → every future visual on-brand, forever
> iridescent card with hover glow? one prompt. done
> opus 4.7 under the hood. separate weekly limit. doesn't touch chat or code
> export to canva. share. tweak. move on
> 2 hours in figma → 2 minutes in claude
> design isn't a skill anymore. it's a prompt.
people who wire this in this week have a month of free leverage on everyone still opening figma.
> Be Saylor
> BTC at $74k. timeline skeptical and scared
> open the ATM. don't flinch.
> issue 21.8M shares of STRC preferred. raise $2.18B.
> issue 2.16M shares of MSTR common. raise another $366M
> leave STRF, STRK, STRD untouched this week. those are for later
> take the full $2.54B. buy 34,164 BTC at avg $74,395
> stack now: 815,061 BTC. cost basis $61.56B. BTC yield 9.5% YTD
> check the remaining capacity
> STRC: $19.5B left. MSTR: $26.7B after march's $21B raise. STRF: $1.6B. STRK: $2.1B. STRD: $4.0B.
> $53B of dry powder still loaded
> 20x this week's buy sitting in the chamber.
> retail: "he'll blow up any day now." ( let toxic idiots like runnerxbt comment retard )
> you: file the 8-K and buy again next monday
> you're not trading bitcoin. you're running a perpetual preferred stock-for-bitcoin engine.
> the float is the fuel. the preferreds are the turbine. btc is the output.
This is peak crypto alpha stack 2026..
Asked my Browser agent one question:
"should i ape $HYPE right now?"
[ Full prompt below ]
watch what it did in 10 minutes..
> Scraped the entire hyperliquid leaderboard
> opened every top trader's profile
> pulled their live positions, entries, stops
> cross-referenced with whale wallets on hypurrscan
> checked funding rates vs binance + okx for arb
> mapped liquidation cascade zones on coinglass
> scanned x for every KOL take in last 24h
then synthesized it all into one research doc with a buy/sell call.
no API exists for half of this. every scrape is done through a real chrome browser
This is peak crypto alpha stack 2026..
Asked my Browser agent one question:
"should i ape $HYPE right now?"
[ Full prompt below ]
watch what it did in 10 minutes..
> Scraped the entire hyperliquid leaderboard
> opened every top trader's profile
> pulled their live positions, entries, stops
> cross-referenced with whale wallets on hypurrscan
> checked funding rates vs binance + okx for arb
> mapped liquidation cascade zones on coinglass
> scanned x for every KOL take in last 24h
then synthesized it all into one research doc with a buy/sell call.
no API exists for half of this. every scrape is done through a real chrome browser
Very important for BTC to hold above the 100D EMA here for few days..
last time bitcoin failed to do so.. it went from 116K to 80K and 97K to 60K
so yeah, If we can't flip this into support and hold above it for a few days, history says this gets ugly fast
Dune analytics MCP.. Claude becomes your on chain SQL analyst..
No dashboard has every query you'll ever need.
dune does but writing SQL is a skill, and most of you would skip it.. i know
this MCP fixes that.
Claude writes the query, runs it on dune, and explains what the data actually means.
with this MCP wired in, you don't need to know SQL.
you describe what you want in plain english and Claude does the rest. like
"Claude, which wallets bought $RAVE in last few weeks and still hold?"
"Claude, show me the top 50 ETH wallets by stablecoin inflows last 7 days."
"Claude, what's the median gas paid by ARB users in the last 24h?"
Questions no dashboard can answer. one prompt away..
Setup (3 minutes)
▫️Step 1: grab a free dune API key → https://t.co/ZtboPrei4j
▫️Step 2: add this to ~/.claude/settings.json or .mcp.json:
{
"mcpServers": {
"dune": {
"command": "npx",
"args": ["-y", "dune-analytics-mcp"],
"env": {
"DUNE_API_KEY": "your-key-here"
}
}
}
}
▫️Step 3: restart claude. you'll see the dune tool load in your tool menu.
that's it. you now have onchain SQL on tap.
How to actually use it:
3 prompts i use:
1) Smart money watchlist:
"claude, pull the top 20 wallets by realized pnl on $TOKEN in the last 30 days. show me which ones are still holding."
gives you a clean leaderboard of who's actually winning on that token. add them to your etherscan watchlist.
2) accumulation vs distribution
"claude, compare net inflows vs outflows for $TOKEN across all CEX wallets in the last 14 days."
if whales are moving off exchanges → accumulation. onto exchanges → distribution. you see the rotation before the candle.
3) narrative heat check
"claude, which 10 tokens saw the biggest % increase in unique new holders this week?"
finds where fresh money is flowing.
before this MCP i'd either, pay for a pro dune account + write queries manually, or look at someone else's dashboard and hope it answers my question…
now claude writes it for me, in seconds, custom to my thesis. no dashboard in existence beats that.
free tier covers most of what you need. upgrade if you're querying heavy.
( built a quick $RAVE post mortem dashboard using a prompt as shown in the video )
MORE SUCH USEFUL MCP for traders below.. 👇
most people using claude for trading are still copy-pasting addresses and asking questions.
the real unlock is wiring in MCPs:
onchain data market feeds prediction markets memory + reasoning
claude stops being a chatbot. starts being a research desk that never sleeps.