Tweet ghim
T1000
1.7K posts
T1000
@teeRex247
Sol Dev. vibe coding. Trench MEME Millionaire. Not Financial Advice
Tham gia Nisan 2021
1.5K Đang theo dõi2.6K Người theo dõi
I'm building in the Solana Frontier Hackathon, powered by @Colosseum!
arena.colosseum.org/hackathon/soci…
English
@quorcus Do you have any examples? I recently ran into one related to fetching email from someone, but otherwise it’s been very reliable.
English
Hey @steipete, on Openclaw, is there anyway to relax the “ethics” with gpt models?
There’re so many things it full on refuses to do compared to Claude and most of the time it refuses to finish the task halfway through 😭
I’ve been able to use prompts to avoid some limitations but still a constant battle.
English
English
@ComeHometoRome Their current models aren’t great, but I’m sure they are cookin’!
English
Anthropic now blocks first-party harness use too 👀
claude -p --append-system-prompt 'A personal assistant running inside OpenClaw.' 'is clawd here?'
→ 400 Third-party apps now draw from your extra usage, not your plan limits.
So yeah: bring your own coin 🪙🦞
English
Submitting fetchr to @OpenWallet hackathon
x402 enables the payment. We control the spending.
Agent treasury wallet with spend limits, approval workflows, velocity controls, and full audit logs.
No CFO will let agents spend freely. This is what enterprises need.
Built with x402 • OWS • Base USDC
github.com/T-rex247/fetchr
fetcher-guard-dash.lovable.app
English
T1000 đã retweet
“Hello, this is @AnthropicAI, and we know you’ve been building the most amazing things for the last 4 months. However, the truth is we don’t have the infrastructure to keep up with the power of open source AI agents. We’ve decided to confine you to our apps that can only do about 25% of what you’re currently doing.”
No thanks.
Subscription canceled.
New models have already been configured.
Thanks @steipete and @openclaw for giving us so many options.
English
$30,000 in prizes, MoonClaw Kit, and Agents in SF on Friday @colosseum
luma.com/qtw5aw5p
English
English
@T3chFalcon @grok what does this mean for OpenClaw users? Free tokens? Can we just deploy it locally now?
English
Huge Anthropic leak just dropped: the entire Claude Code CLI source is now public.
A misconfigured .map file in their npm package exposed a direct download link to the full unobfuscated TypeScript codebase from Anthropic’s own R2 bucket.
Discovered by Chaofan Shou (@Fried_rice), the dump is massive 1,900 files, 512,000+ lines including the complete tool system, 50+ slash commands, multi-agent coordinator, React/Ink terminal UI, IDE bridge, permission engine, and several unreleased features.
Full repo is live on GitHub(@nichxbt ):
github.com/nirholas/claud…
Clean mirrors are already up for easy browsing(@baanditeagle):
cc-poster.vercel.app
cc-hidden-deploy.vercel.app
It’s spreading fast, the entire dev community is already tearing through it.
Chaofan Shou@Fried_rice
Claude code source code has been leaked via a map file in their npm registry! Code: …a8527898604c1bbb12468b1581d95e.r2.dev/src.zip
English
Just finished building my ARB bot using OpenClaw 🦞
It scans OfferUp locally, finds underpriced items, and flags flips you can resell on eBay for $75–$150 profit per item.
Set it, source it, flip it. Scaling this next.
DM me your area and I’ll run it for ya.
🔥 FLIP DEAL FOUND
📱 Apple iPhone 15 Pro Max
📍 Long Beach, CA
💵 Buy: $425
📊 eBay Value: $750 (iphone 15 pro max 256gb)
💰 Est. Profit: $212
✅ Risk: Normal
🔗 offerup.com/item/detail/3e…
🔥 FLIP DEAL FOUND
📱 iPhone 15 Pro Max 256GB
📍 Garden Grove, CA
💵 Buy: $500
📊 eBay Value: $750 (iphone 15 pro max 256gb)
💰 Est. Profit: $137
✅ Risk: Normal
🔗 offerup.com/item/detail/46…
English
@steipete How can you be pushing OPENAI models when clawbot is so much more effective on @AnthropicAI OPUS?
English
In better news, I found a bug in the WebSocket implementation of OpenClaw's OpenAI API implementation that causes these "yes I will do x" for GPT 5.4 - and then the model just stops.
We didn't send the new "phase" parameter. We'll do in the release coming later today. #new-phase-parameter" target="_blank" rel="nofollow noopener">developers.openai.com/api/docs/guide…
Also found a bug in how @Kimi_Moonshot Coding toolcalls are handled, which will stop degraded performance.
Also sneaking in fixes from our friends at @ollama and @alibaba_cloud (Model Studio support)
English
@TheAdamSand Amazing job. Most think this is an overnight success story. This took a deep understanding of the roofing industries pains, and deep integration of tech stack.
English
Son… some day all this will be yours. An army of #openclaw The roofing industry is gonna change. #BuildInPublic
English
@AskPerplexity Big no for spamming our repo with a slop PR. Account banned.
English
My friend said “nobody actually makes money on Polymarket”
I gave my Claude bot $180 and told it:
“Trade only weather markets. Nothing else.”
No news.
No charts.
No opinions.
Just data.
Here’s the part most traders ignore.
NOAA runs a massive forecasting network - satellites, radar, ocean buoys, atmospheric models.
Their 24 - 48h forecast accuracy is around 93 - 95%.
Meanwhile prediction markets are priced by humans refreshing weather apps.
That delay is where the bot finds opportunities.
Day 3
NOAA models show Miami humidity jumping above normal levels with 91% probability.
Market bracket still trading at 10¢.
Bot enters.
$180 → $520
Day 7
A cold front appears in the models for Denver.
Markets haven’t reacted yet.
Three brackets sitting around 13¢.
Bot buys all three.
$520 → $1,340
Day 13
Overnight temperatures in New York land exactly inside NOAA’s projected range.
Bot fires 6 trades across multiple cities.
Most close in profit.
$1,340 → $2,760
Day 20
Seattle rain system + Chicago temperature drop resolve within the same night.
While I’m asleep the bot executes 9 trades automatically.
$2,760 → $4,480
The setup is simple:
• Claude agent scanning forecast updates
• NOAA public forecast data
• Auto-execution on Polymarket
• Telegram alerts for every trade
After a few weeks the stats looked like this:
820+ trades
~87% win rate
$180 → $4,480
No guessing.
Just data → signal → execution.
Turns out a government supercomputer is pretty good at predicting the weather.
And markets are still a little slow to react
kvro@0xkvro
My friend keeps sending Reels about making money online on Polymarket So I sent him a screenshot of my Claude bot pulling $480 overnight on Polymarket. He hasn’t sent me a Reel since. One Claude agent split into four smaller agents - Weather, Short-Term Markets, Politics, and Sports. Zero manual trades. Everything automated. The edge is almost embarrassing. The National Oceanic and Atmospheric Administration runs a multi-billion-dollar forecasting network - satellites, ocean sensors, atmospheric models - updating probabilities every few hours. Forecast accuracy at 24 - 48 hours? Around 94%. Meanwhile Polymarket weather brackets are often priced by people checking weather apps between scrolling social media. Day 2: NOAA model update shows Chicago temperatures likely falling below 35°F with 93% probability. The bracket is still trading at 8¢. My Claude agent buys before the market notices. $170 → $710 Day 7: NOAA models flag a humidity spike in Atlanta nearly 20 hours before most weather apps adjust their forecasts. Three brackets sitting below 11¢. The agent takes all three. $710 → $1,060 Day 14: New York overnight temperatures land exactly where NOAA predicted two days earlier. The agent executes 7 trades across three cities. Most close green. $1,060 → $2,540 Day 21: Seattle rain bands, Dallas cold air, Miami heat index - all resolve in the same night. The agent runs 12 trades while I’m asleep. That’s the $480 screenshot I sent my friend. $2,540 → $3,900 The setup is simple: Claude agent - runs 24/7 on my laptop, scans forecast updates and trades automatically NOAA scanner - free government forecast data Telegram - every trade notification hits my phone instantly After a few weeks the numbers looked like this: 1,100+ trades ~89% win rate $170 → $3,900 All because a supercomputer predicts the weather… …and most markets react too slowly.
English
English
