docs: update README for v2.5 - Polymarket + HN as killer features, add Anthropic odds example
Reorder v2.5 features: Polymarket prediction markets and HN lead as #1, multi-signal quality-ranked relevance scoring as #2. Add Anthropic Odds example showcasing 11 live markets from a two-word query. Add Anthropic and OpenAI Polymarket transcripts to launch tweets. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Matt Van Horn
+488
-1
@@ -247,4 +247,491 @@ Key findings:
|
||||
- People building on top - @tjarkoleifer created "re-skill" meta skill, @rajachirravuri recommends it as part of a PM stack
|
||||
- Coverage: Alejandro AO crash course (39K views, 1,049 likes), Jason Calacanis on This Week in Startups (24K views)
|
||||
- 1.5K GitHub stars, listed on skills.sh and Smithery
|
||||
- Grok itself correctly attributed the skill when asked about ithah
|
||||
- Grok itself correctly attributed the skill when asked about ithah
|
||||
|
||||
### Arizona Basketball (GENERAL) - VERIFIED 2/26, POLYMARKET INTEGRATION
|
||||
|
||||
Stats: 2 Reddit threads | 112 upvotes | 37 X posts | ~4,200 likes | ~380 reposts | 15 YouTube videos | 517K views | 3 with transcripts | 6 Polymarket markets | ~30 web pages
|
||||
Top voices: @ArizonaMBB (960 likes), @MrAzSports (449 likes), @JonRothstein | r/CollegeBasketball
|
||||
|
||||
Key findings:
|
||||
- Arizona 25-2 (13-2 Big 12), set program record with 22-0 start, 2-game Big 12 lead with 3 games left
|
||||
- Projected #1 seed in West region (San Jose) for March Madness per NCAA.com and CBS bracket previews
|
||||
- Polymarket odds: Championship 13%, #1 Seed 88%, Big 12 title race Arizona leads by 2
|
||||
- The Field of 68 dubbed them "the TOUGHEST team in America" after escaping Baylor 87-80 shorthanded
|
||||
- #1 ranking debate: Duke vs Arizona vs Michigan - Jon Rothstein has Duke #1, Arizona #2
|
||||
- Kansas rematch Feb 28: @ArizonaMBB promoting "STRIPE OUT" at McKale Center, first meeting highlight video 248K views on ESPN YouTube
|
||||
- Tommy Lloyd's March Madness record looms: 6-4 in four NCAA trips, never past Sweet 16
|
||||
|
||||
Note: This was the test that validated the Polymarket two-pass query expansion (v2.2-cross). Previous version only found 2 markets (Big 12 title + irrelevant NAU game). New version finds 6 markets including Championship odds, #1 Seed, and Big 12 title race via tag-based domain expansion.
|
||||
|
||||
Full output:
|
||||
|
||||
```
|
||||
▐▛███▜▌ Claude Code v2.1.59
|
||||
▝▜█████▛▘ Opus 4.6 · Claude Max
|
||||
▘▘ ▝▝ /Users/mvanhorn
|
||||
|
||||
❯ /last30daysCROSS arizona basketball
|
||||
|
||||
⏺ I'll research arizona basketball across Reddit, X, and the web to find what's been discussed in the last 30
|
||||
days.
|
||||
|
||||
Parsed intent:
|
||||
- TOPIC = arizona basketball
|
||||
- TARGET_TOOL = unknown
|
||||
- QUERY_TYPE = GENERAL
|
||||
|
||||
Research typically takes 2-8 minutes. Starting now.
|
||||
|
||||
⏺ Web Search("Arizona Wildcats basketball X twitter handle site:x.com")
|
||||
⎿ Did 1 search in 3s
|
||||
|
||||
⏺ Resolved X handle: @ArizonaMBB (official Arizona Basketball account). Running research now.
|
||||
|
||||
⏺ Bash(for dir in \
|
||||
"." \…)
|
||||
⎿ /last30days · researching: arizona basketball
|
||||
┌─────────────────────────────────────────────────────┐
|
||||
│ /last30days v2.1 — Source Status │
|
||||
… +129 lines (ctrl+o to expand)
|
||||
⎿ (timeout 5m)
|
||||
|
||||
⏺ Research script complete. Now supplementing with web search.
|
||||
|
||||
⏺ Web Search("arizona wildcats basketball 2026 season")
|
||||
⎿ Did 1 search in 7s
|
||||
|
||||
⏺ Web Search("arizona basketball NCAA tournament March Madness 2026")
|
||||
⎿ Did 1 search in 6s
|
||||
|
||||
⏺ Web Search("arizona wildcats basketball rankings Big 12 February 2026")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ What I learned:
|
||||
|
||||
Arizona is having a historic season and the narrative is "Final Four or bust." The Wildcats are 25-2
|
||||
(13-2 Big 12), set a program record with a 22-0 start, and hold a 2-game lead atop the Big 12 with three
|
||||
games left. They're projected as a #1 seed in the West region (San Jose) for March Madness, per NCAA.com
|
||||
and CBS bracket previews.
|
||||
|
||||
The team's identity is toughness and balance. The Field of 68 dubbed them "the TOUGHEST team in America"
|
||||
after they escaped Baylor 87-80 on Feb 24 - and they did it shorthanded, without Koa Peat or Dwayne
|
||||
Aristode. CBS Sports highlighted how "Arizona's BALANCE could take them to a Final Four" after the Houston
|
||||
road win. @ArizonaMBB's postgame content from the Baylor win pulled nearly 1,000 likes per post.
|
||||
|
||||
The #1 ranking debate is real - Duke, Arizona, or Michigan? Jon Rothstein's latest rankings have Duke #1,
|
||||
Arizona #2, Michigan #3, per @JonRothstein. The Field of 68 ran a whole segment asking "Who MOST deserves
|
||||
to be No. 1?" with the verdict being "All THREE have a case." On Polymarket, bettors give Arizona a 13%
|
||||
chance to win the NCAA Tournament (Duke and Michigan each at 18%) and an 88% chance to earn a #1 seed.
|
||||
|
||||
Saturday's Kansas rematch is the game everyone's watching. @ArizonaMBB is promoting a "STRIPE OUT" for the
|
||||
Feb 28 home game vs KU. Arizona beat Kansas 86-83 in Lawrence on Feb 10 (that highlight video has 248K
|
||||
views on ESPN's YouTube), and Locked On Wildcats is already breaking down "Arizona's ADJUSTMENTS to BEAT
|
||||
KANSAS" with a focus on Darryn Peterson's role. Nebraska fans are even rooting for Arizona in this one, per
|
||||
@jamespruch.
|
||||
|
||||
The Big 12 title race is Arizona's to lose, but the schedule is brutal. Kansas, Iowa State, Houston, and
|
||||
Texas Tech are all tied at 11-4, two games back. Arizona closes with Kansas (Feb 28) and Iowa State - the
|
||||
toughest remaining schedule in the conference, per Heartland College Sports. A four-way tie for second is
|
||||
still possible.
|
||||
|
||||
Tommy Lloyd's March Madness record looms large. The Wildcats are 6-4 in four NCAA Tournament trips under
|
||||
Lloyd and haven't made it past the Sweet 16. Yahoo Sports is asking the "Final Four or bust?" question
|
||||
directly - the talent and record demand a deep run this year.
|
||||
|
||||
---
|
||||
All agents reported back!
|
||||
├─ 🟠 Reddit: 2 threads │ 112 upvotes
|
||||
├─ 🔵 X: 37 posts │ ~4,200 likes │ ~380 reposts
|
||||
├─ 🔴 YouTube: 15 videos │ 517K views │ 3 with transcripts
|
||||
├─ 📊 Polymarket: 6 markets │ Championship: 13%, #1 Seed: 88%, Big 12 title race: Arizona leads by 2
|
||||
├─ 🌐 Web: ~30 pages (supplementary)
|
||||
└─ 🗣️ Top voices: @ArizonaMBB (960 likes), @MrAzSports (449 likes), @JonRothstein │ r/CollegeBasketball
|
||||
---
|
||||
|
||||
---I'm now an expert on Arizona basketball. Some things I can help with:
|
||||
- Break down the Kansas rematch - what adjustments does Arizona need and what's Darryn Peterson's role?
|
||||
- Is "Final Four or bust" fair given Tommy Lloyd's March track record?
|
||||
- Compare Arizona's title odds vs Duke and Michigan - who has the best path through the bracket?
|
||||
|
||||
✻ Sautéed for 3m 51s
|
||||
```
|
||||
|
||||
### Iran War (NEWS) - VERIFIED 2/26, POLYMARKET INTEGRATION
|
||||
|
||||
Stats: 7 Reddit threads | 439 upvotes | 30 X posts | ~30 likes | 20 YouTube videos | 2,030,730 views | 5 with transcripts | 18 HN stories | 1,187 points | 849 comments | 14 Polymarket markets | $425M+ volume | 30 web pages
|
||||
Top voices: Breaking Points (435K views), Judge Napolitano/Scott Ritter (272K views), TBN Israel (263K views) | r/politics, r/SubredditDrama
|
||||
|
||||
Key findings:
|
||||
- Simultaneous war buildup and diplomacy: 150+ aircraft, two carrier strike groups deployed while third round of Geneva nuclear talks happening today
|
||||
- Polymarket odds with $425M in volume: US strikes by 2026 70%, War Powers invocation 60%, Israel strikes by June 64%, Khamenei out by March 18%, enrichment deal 16%, formal war declaration only 8%
|
||||
- Pentagon brass leaking opposition hard - Breaking Points (435K views) covered leaked assessments showing military leadership doesn't want this
|
||||
- Scott Ritter's war breakdown on Judge Napolitano pulled 272K views, 15.9K likes
|
||||
- Iraq War 2003 parallels drawn across all platforms - Al Jazeera ran direct comparison piece
|
||||
- Iran playing both sides: FM Araghchi says deal "within reach" while IRGC seized two tankers near Farsi Island
|
||||
- r/Conservative "imploding" after Trump struck Iran - "no wars" promise colliding with reality per r/SubredditDrama
|
||||
- Markets expect limited strikes with War Powers, NOT a declared war - the sharpest signal in the entire research
|
||||
|
||||
Note: This was the "perfect" Polymarket test case - 14 markets found via tag-based domain expansion ("Geopolitics", "Middle East"). Iran/War appear directly in event titles so first-pass queries already find most markets. The domain expansion added 4 more.
|
||||
|
||||
Full output:
|
||||
|
||||
```
|
||||
▐▛███▜▌ Claude Code v2.1.59
|
||||
▝▜█████▛▘ Opus 4.6 · Claude Max
|
||||
▘▘ ▝▝ /Users/mvanhorn
|
||||
|
||||
❯ /last30daysCROSS iran war
|
||||
|
||||
⏺ I'll research iran war across Reddit, X, YouTube, Hacker News, Polymarket, and the web
|
||||
to find what's been discussed in the last 30 days.
|
||||
|
||||
Parsed intent:
|
||||
- TOPIC = iran war
|
||||
- TARGET_TOOL = unknown
|
||||
- QUERY_TYPE = NEWS
|
||||
|
||||
Research typically takes 2-8 minutes. Starting now.
|
||||
|
||||
⏺ Bash(# Find skill root and run research script
|
||||
for dir in \…)
|
||||
⎿ (No output)
|
||||
⎿ (timeout 5m)
|
||||
|
||||
⏺ The script returned no output. Let me check what happened and try again with more
|
||||
diagnostics.
|
||||
|
||||
⏺ Bash(SKILL_ROOT="$HOME/.claude/skills/last30daysCROSS"
|
||||
[ -f "$SKILL_ROOT/scripts/last30days.py" ] && echo "Found at $SKILL_ROOT" || echo
|
||||
"Not found at $SKILL_ROOT"…)
|
||||
⎿ Not found at ls -la /scripts/
|
||||
|
||||
⏺ Bash(for dir in \
|
||||
"$HOME/.claude/skills/last30daysCROSS" \…)
|
||||
⎿ FOUND: /Users/mvanhorn/.claude/skills/last30daysCROSS/scripts/last30days.py
|
||||
FOUND: /Users/mvanhorn/.claude/skills/last30days/scripts/last30days.py
|
||||
FOUND: /Users/mvanhorn/.agents/skills/last30days/scripts/last30days.py
|
||||
FOUND: /Users/mvanhorn/.codex/skills/last30days/scripts/last30days.py
|
||||
|
||||
⏺ Bash(python3 "$HOME/.claude/skills/last30daysCROSS/scripts/last30days.py" "iran war"
|
||||
--emit=compact 2>&1)
|
||||
⎿ /last30days · researching: iran war
|
||||
┌─────────────────────────────────────────────────────┐
|
||||
│ /last30days v2.1 — Source Status │
|
||||
… +287 lines (ctrl+o to expand)
|
||||
⎿ (timeout 5m)
|
||||
|
||||
⏺ Good, got extensive results. Now let me supplement with web searches.
|
||||
|
||||
⏺ Web Search("iran war news February 2026")
|
||||
⎿ Did 1 search in 11s
|
||||
|
||||
⏺ Web Search("US Iran military strikes update 2026")
|
||||
⎿ Did 1 search in 10s
|
||||
|
||||
⏺ Web Search("Iran nuclear deal negotiations Trump 2026")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ What I learned:
|
||||
|
||||
The US-Iran crisis is at a knife's edge - simultaneous war buildup and diplomacy. The
|
||||
US has deployed its largest military concentration in the Middle East in decades -
|
||||
150+ aircraft, two carrier strike groups (Abraham Lincoln and Gerald R. Ford) - while
|
||||
a third round of nuclear talks is literally happening today in Geneva, per NPR and
|
||||
Axios. Polymarket bettors have put $425M on strike timing, with 70% odds of US strikes
|
||||
by year-end and 57% by March 31.
|
||||
|
||||
Pentagon brass is leaking opposition hard. Top generals are warning through back
|
||||
channels that an Iran war would be a "disaster" - Breaking Points covered leaked
|
||||
Pentagon assessments showing military leadership desperately doesn't want this. On
|
||||
YouTube, the Breaking Points video alone pulled 435K views. Scott Ritter's breakdown
|
||||
on Judge Napolitano (272K views, 15.9K likes) laid out what a war would actually look
|
||||
like operationally. Per @TheGreenElf29 and r/politics, Iraq War parallels are
|
||||
everywhere - Al Jazeera ran a piece directly comparing Trump's 2026 playbook to Bush's
|
||||
2003 script.
|
||||
|
||||
Iran is playing both sides - "deal within reach" while seizing tankers. FM Araghchi
|
||||
said a deal is "within reach" if diplomacy gets priority, but Iran's IRGC seized two
|
||||
oil tankers near Farsi Island in early February and attempted to intercept a
|
||||
US-escorted tanker in the Strait of Hormuz. Iran's line: any US strike, even
|
||||
"limited," would be considered an act of aggression requiring retaliation, per
|
||||
r/politics and Al Jazeera.
|
||||
|
||||
Prediction markets are the sharpest signal here. The money says:
|
||||
- US strikes Iran by 2026: 70% (down 33% this month as talks progressed)
|
||||
- Trump invokes War Powers by March 31: 60% (up 19% this month)
|
||||
- Israel strikes Iran by June: 64% (up 8%)
|
||||
- Khamenei out by March 31: 18%
|
||||
- Iran agrees to end enrichment by March 31: 16%
|
||||
- Official US war declaration: only 8%
|
||||
|
||||
The takeaway: markets expect strikes but NOT a declared war - a limited strike
|
||||
scenario with War Powers invocation is the consensus bet.
|
||||
|
||||
Congress is pushing back. Sen. Kaine is pushing a war powers resolution requiring a
|
||||
congressional vote before any Iran strike. On YouTube, his statement got 10K views.
|
||||
Reddit's r/politics had three threads in the top-scored results all framing war as
|
||||
imminent and catastrophic. Notably, a meta-thread on r/NoStupidQuestions (70 upvotes)
|
||||
asked whether Reddit was censoring Iran war discussions - suggesting the topic is
|
||||
hitting mainstream awareness fast.
|
||||
|
||||
r/Conservative is splitting. A SubredditDrama thread documented r/Conservative
|
||||
"imploding" after Trump struck Iran - the "no wars" promise colliding with reality,
|
||||
per r/SubredditDrama.
|
||||
|
||||
KEY PATTERNS from the research:
|
||||
1. Dual-track crisis - war preparation and diplomacy happening simultaneously, with
|
||||
Trump setting a 10-15 day deadline for a "meaningful deal," per Axios
|
||||
2. Iraq playbook echo - commentators across all platforms drawing direct 2003
|
||||
parallels, per Al Jazeera and r/politics
|
||||
3. Military doesn't want this - Pentagon leaking opposition is a strong signal of
|
||||
internal resistance, per Breaking Points (435K views)
|
||||
4. Strait of Hormuz as flashpoint - tanker seizures and Iran's submarine capability in
|
||||
shallow waters making this the likeliest escalation trigger, per HN and Stars and
|
||||
Stripes
|
||||
5. Real money says limited strikes, not full war - Polymarket odds suggest strikes but
|
||||
only 8% chance of formal war declaration
|
||||
|
||||
---✅ All agents reported back!
|
||||
├─ 🟠 Reddit: 7 threads │ 439 upvotes
|
||||
├─ 🔵 X: 30 posts │ ~30 likes │ mostly real-time commentary
|
||||
├─ 🔴 YouTube: 20 videos │ 2,030,730 views │ 5 with transcripts
|
||||
├─ 🟡 HN: 18 stories │ 1,187 points │ 849 comments
|
||||
├─ 📊 Polymarket: 14 markets │ US strikes by 2026: 70%, War Powers: 60%, Israel
|
||||
strikes by June: 64%, Khamenei out March: 18%, Enrichment deal: 16%
|
||||
├─ 🌐 Web: 30 pages (supplementary)
|
||||
└─ 🗣️ Top voices: Breaking Points (435K views), Judge Napolitano/Scott Ritter (272K
|
||||
views), TBN Israel (263K views) │ r/politics, r/SubredditDrama
|
||||
---
|
||||
I'm now an expert on the Iran war situation. Some things you could ask:
|
||||
- What are the realistic scenarios for the next 2 weeks given the Geneva talks
|
||||
deadline?
|
||||
- How does the military buildup compare to June 2025 strikes, and what would "limited
|
||||
strikes" actually target?
|
||||
- What happens to oil markets and global economy if the Strait of Hormuz gets shut
|
||||
down?
|
||||
|
||||
✻ Cogitated for 6m 5s
|
||||
```
|
||||
|
||||
### Anthropic Odds (PREDICTION MARKETS) - VERIFIED 2/26, POLYMARKET INTEGRATION
|
||||
|
||||
Stats: 25 X posts | 218 likes | 38 reposts | 13 YouTube videos | 719,854 views | 4 with transcripts | 6 HN stories | 48 points | 36 comments | 11 Polymarket markets | ~30 web pages
|
||||
Top voices: @tradermap_whale (whale trades), @predictheory (IPO odds), @Lolipeterh (Pentagon analysis) | CBS Mornings, CNN, Bloomberg
|
||||
|
||||
Key findings:
|
||||
- Pentagon standoff is THE story: Defense Secretary Hegseth gave Anthropic a Friday deadline to drop AI guardrails for military use or face blacklisting via Defense Production Act
|
||||
- Polymarket: Best AI model Feb 98% (Anthropic), March 61%, IPO before OpenAI 64%, $500B+ valuation 87%, FrontierMath 50% score 48% (up 28% today)
|
||||
- Anthropic quietly loosened safety policy amid Pentagon pressure - CNN reports a caveat added to core safety promise. HN top comment: "The reality of competition."
|
||||
- IPO odds volatile: Kalshi IPO-first odds "fell through the floor, ~70% down to the low teens in one move" per @predictheory
|
||||
- Pentagon ban odds only 22% - money says it's posturing, not real blacklisting
|
||||
|
||||
Note: Perfect showcase for Polymarket as a source - 11 markets spanning model benchmarks, IPO timing, valuation milestones, Pentagon politics, and math benchmarks. Two-pass expansion found "Anthropic" as an outcome inside broader "best AI model" and "AI company IPO" events. Just two words ("anthropic odds") produced a comprehensive prediction market intelligence briefing.
|
||||
|
||||
Full output:
|
||||
|
||||
```
|
||||
▐▛███▜▌ Claude Code v2.1.59
|
||||
▝▜█████▛▘ Opus 4.6 · Claude Max
|
||||
▘▘ ▝▝ /Users/mvanhorn
|
||||
|
||||
❯ /last30daysCROSS anthropic odds
|
||||
|
||||
⏺ I'll research "anthropic odds" across Reddit, X, and the web to find what's been
|
||||
discussed in the last 30 days.
|
||||
|
||||
Parsed intent:
|
||||
- TOPIC = anthropic odds
|
||||
- TARGET_TOOL = unknown
|
||||
- QUERY_TYPE = GENERAL (broad interest in Anthropic's prospects, prediction markets,
|
||||
competitive positioning)
|
||||
|
||||
Research typically takes 2-3 minutes. Starting now.
|
||||
|
||||
⏺ Bash(for dir in \
|
||||
"." \…)
|
||||
⎿ /last30days · researching: anthropic odds
|
||||
┌─────────────────────────────────────────────────────┐
|
||||
│ /last30days v2.1 — Source Status │
|
||||
… +221 lines (ctrl+o to expand)
|
||||
⎿ (timeout 5m)
|
||||
|
||||
⏺ Now let me run supplementary web searches.
|
||||
|
||||
⏺ Web Search("Anthropic odds prediction market 2026")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ Web Search("Anthropic IPO odds valuation 2026")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ Web Search("Anthropic Pentagon standoff AI military 2026")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ What I learned:
|
||||
|
||||
Pentagon standoff is THE story right now - Defense Secretary Hegseth gave Anthropic a
|
||||
Friday deadline to drop AI guardrails for military use or face blacklisting via the
|
||||
Defense Production Act, per CBS, CNN, Bloomberg, and a wave of X posts from @The__GDD,
|
||||
@trendy_tech_, and @jimkaskade. The trigger was Claude's use during the Maduro
|
||||
capture operation in January. Anthropic CEO Dario Amodei insists AI-controlled weapons
|
||||
and mass surveillance are lines the company won't cross. Polymarket traders put the
|
||||
ban odds at just 22%, signaling they think it's posturing, per @Lolipeterh.
|
||||
|
||||
Anthropic loosened its safety policy anyway - Amid the pressure, CNN reports Anthropic
|
||||
quietly added a caveat to its core safety promise. HN's @KerrickStaley posted the
|
||||
Bloomberg piece on this, with the top comment simply: "The reality of competition."
|
||||
This is a significant moment - the first crack in Anthropic's safety-first brand.
|
||||
|
||||
Prediction markets love Anthropic's tech, skeptical on IPO - Polymarket gives
|
||||
Anthropic a 98% chance of having the best AI model at end of February and 61% for
|
||||
March (Google at 22%, OpenAI at 10%). Claude 4.6 is dominating. But the IPO picture is
|
||||
murkier: @predictheory flagged that Anthropic IPO-first odds on Kalshi "fell through
|
||||
the floor, ~70% down to the low teens in one move." Polymarket has Anthropic at 64% to
|
||||
IPO before OpenAI, but 95% NO on an IPO by June 2026. Meanwhile, 87% odds Anthropic
|
||||
hits $500B+ valuation this year - current valuation is $380B after a $30B raise led by
|
||||
GIC and Coatue, per Fortune.
|
||||
|
||||
Claude FrontierMath odds surging - Polymarket's "Will Claude score 50% on
|
||||
FrontierMath?" market jumped 28% today to 48% YES. This is a live bet on whether
|
||||
Claude can crack elite-level math benchmarks by June 30.
|
||||
|
||||
Big money is piling in - Dan Sundheim's D1 Capital bets on Anthropic featured on
|
||||
Invest Like The Best (18K views), and "Amazon's Quiet Bet on Anthropic" from Dumb
|
||||
Money Live got 9K views. Anthropic's Super Bowl ads and the @grok account noting
|
||||
Claude "self-reporting 15-20% consciousness odds" added to the buzz.
|
||||
|
||||
---
|
||||
All agents reported back!
|
||||
├─ 🟠 Reddit: 3 threads (older, low engagement - all outside 30-day window)
|
||||
├─ 🔵 X: 25 posts │ 218 likes │ 38 reposts
|
||||
├─ 🔴 YouTube: 13 videos │ 719,854 views │ 4 with transcripts
|
||||
├─ 🟡 HN: 6 stories │ 48 points │ 36 comments
|
||||
├─ 📊 Polymarket: 11 markets │ Best model Feb: 98%, March: 61%, IPO first: 64%, $500B+
|
||||
val: 87%, FrontierMath 50%: 48% (up 28% today)
|
||||
├─ 🌐 Web: ~30 pages (supplementary)
|
||||
└─ 🗣️ Top voices: @tradermap_whale (whale trades), @predictheory (IPO odds),
|
||||
@Lolipeterh (Pentagon analysis) │ CBS Mornings, CNN, Bloomberg
|
||||
---
|
||||
|
||||
---I'm now an expert on Anthropic odds. Some things I can help with:
|
||||
- What are the realistic outcomes of the Pentagon standoff - does Anthropic fold,
|
||||
fight, or find a middle ground?
|
||||
- Break down the prediction market landscape - where's the smart money, and which bets
|
||||
look mispriced?
|
||||
- How does the safety policy loosening change Anthropic's competitive position vs
|
||||
OpenAI and Google?
|
||||
|
||||
✻ Cooked for 2m 44s
|
||||
```
|
||||
|
||||
### OpenAI Polymarket (PREDICTION MARKETS) - VERIFIED 2/26, POLYMARKET INTEGRATION
|
||||
|
||||
Stats: 33 X posts | 1,129+ likes | 90+ reposts | 11 YouTube videos | 350,675 views | 4 with transcripts | 14 Polymarket markets | $21M+ volume (IPO markets alone) | 10 web pages
|
||||
Top voices: @Polymarket (855 likes), @jayendra_jog (59), @camolNFT (36) | r/OpenAI, r/PolymarketTrading
|
||||
|
||||
Key findings:
|
||||
- OpenAI IPO is the mega-market: $21M+ wagered, 51.5% odds of IPO by end of 2026, leading closing cap $1.25T-$1.5T at 52%
|
||||
- AI agents eating Polymarket alive: automated traders extracted ~$40M through arbitrage, windows compressed from 12.3s to 2.7s
|
||||
- OpenAI hardware bets: 200+ people working on "AI lamp", only 3.2% chance of consumer hardware by March, earbuds most likely at 46%
|
||||
- Anthropic winning model race: 98% best model Feb, OpenAI "not even competitive this month"
|
||||
- GPT-5.3 mystery bet: new wallet dropped $2,500 on NOT releasing by Feb 28, potential "OpenAI insider" per @camolNFT
|
||||
- Sam Altman equity watch: 33% chance he gets OpenAI equity by June 30
|
||||
|
||||
Note: 14 Polymarket markets found - tag-based domain expansion pulled in "AI Companies", "Tech IPO" category markets. The AI agent arbitrage story is a perfect Polymarket meta-narrative: prediction markets being traded by the same AI tools we're researching.
|
||||
|
||||
Full output:
|
||||
|
||||
```
|
||||
▐▛███▜▌ Claude Code v2.1.59
|
||||
▝▜█████▛▘ Opus 4.6 · Claude Max
|
||||
▘▘ ▝▝ /Users/mvanhorn
|
||||
|
||||
❯ /last30daysCROSS openai polymarket
|
||||
|
||||
⏺ I'll research openai polymarket across Reddit, X, and the web to find what's been
|
||||
discussed in the last 30 days.
|
||||
|
||||
Parsed intent:
|
||||
- TOPIC = openai polymarket
|
||||
- TARGET_TOOL = unknown
|
||||
- QUERY_TYPE = GENERAL
|
||||
|
||||
Research typically takes 2-3 minutes. Starting now.
|
||||
|
||||
⏺ Bash(cd "$HOME/.claude/skills/last30daysCROSS" && python3 scripts/last30days.py
|
||||
"openai polymarket" --emit=compact 2>&1)
|
||||
⎿ /last30days · researching: openai polymarket
|
||||
┌─────────────────────────────────────────────────────┐
|
||||
│ /last30days v2.1 — Source Status │
|
||||
… +198 lines (ctrl+o to expand)
|
||||
⎿ (timeout 5m)
|
||||
|
||||
⏺ Good data. Now let me supplement with web searches.
|
||||
|
||||
⏺ Web Search("OpenAI Polymarket prediction markets 2026")
|
||||
⎿ Did 1 search in 7s
|
||||
|
||||
⏺ Web Search("OpenAI IPO prediction market odds discussion")
|
||||
⎿ Did 1 search in 8s
|
||||
|
||||
⏺ Web Search("AI agents trading Polymarket bots 2026")
|
||||
⎿ Did 1 search in 9s
|
||||
|
||||
⏺ What I learned:
|
||||
|
||||
OpenAI's IPO is the mega-market - Polymarket traders have wagered $21M+ across
|
||||
multiple markets tracking OpenAI's path to going public. The market gives just 51.5%
|
||||
odds of an IPO by end of 2026, with "No IPO by Dec 31" at 49%. If it happens, the
|
||||
leading closing cap bracket is $1.25T-$1.5T at 52% among if-IPO outcomes. Meanwhile,
|
||||
only 11% chance of a $1T+ IPO before 2027 - and SpaceX crushes it at 95% vs OpenAI's
|
||||
5% for higher IPO cap, per @Polymarket. The massive $100B funding round at $830B
|
||||
valuation suggests OpenAI may be in no rush, per MLQ AI.
|
||||
|
||||
AI agents are eating Polymarket alive - This is the real crossover story. Automated
|
||||
traders have extracted ~$40M through arbitrage, per Yahoo Finance. Arbitrage windows
|
||||
have compressed from 12.3 seconds (2024) to just 2.7 seconds. OpenClaw and Clawdbot
|
||||
are the dominant agent platforms - Coin Bureau Trading documented +1,560% ROI using
|
||||
OpenClaw for Polymarket trading. New infrastructure like LuckyLobster just launched as
|
||||
a dedicated AI-native execution layer for autonomous Polymarket trading, per
|
||||
@jayendra_jog noting that crypto/prediction markets have "never had more mainstream
|
||||
adoption."
|
||||
|
||||
OpenAI hardware bets are heating up - @Polymarket broke that OpenAI reportedly has
|
||||
200+ people working on an "AI lamp" and other devices (810 likes). But the market is
|
||||
skeptical: only 3.2% chance of a consumer hardware launch by March 31.
|
||||
Earbuds/headphones are the most likely product category at 46%, per @k_shamil57907.
|
||||
|
||||
Anthropic is winning the model race, not OpenAI - The "best AI model end of February"
|
||||
market has Anthropic at 98% with Google down 6.8% this week. OpenAI isn't even
|
||||
competitive this month. On X, @JFahandazh is "very bullish on Anthropic - OpenAI could
|
||||
never be this good."
|
||||
|
||||
GPT-5.3 mystery bet - @camolNFT flagged a new wallet dropping $2,500 on GPT-5.3 NOT
|
||||
releasing by Feb 28, calling it a potential "OpenAI insider." The bet would net $1,000
|
||||
profit if right.
|
||||
|
||||
Sam Altman equity watch - Will Altman get OpenAI equity by June 30? Currently at 33%
|
||||
on Polymarket, with @lexispawn taking a position at those odds, arguing "OpenAI board
|
||||
restructuring is happening NOW."
|
||||
|
||||
---✅ All agents reported back!
|
||||
├─ 🟠 Reddit: 3 threads │ low engagement (all below relevance threshold)
|
||||
├─ 🔵 X: 33 posts │ 1,129+ likes │ 90+ reposts
|
||||
├─ 🔴 YouTube: 11 videos │ 350,675 views │ 4 with transcripts
|
||||
├─ 📊 Polymarket: 14 markets │ IPO cap: 52% at $1.25-1.5T, Best AI model: Anthropic
|
||||
98%, Hardware by March: 3.2%, Altman equity: 33%, GPT-5.3 by Feb 28: betting NO
|
||||
├─ 🌐 Web: 10 pages (supplementary)
|
||||
└─ 🗣️ Top voices: @Polymarket (855 likes), @jayendra_jog (59), @camolNFT (36) │
|
||||
r/OpenAI, r/PolymarketTrading
|
||||
|
||||
I'm now an expert on OpenAI + Polymarket. Some things I can help with:
|
||||
- Break down the IPO timing odds vs. the $100B private funding - is the market saying
|
||||
OpenAI doesn't need to go public?
|
||||
- Explain the AI agent trading ecosystem (OpenClaw vs Clawdbot vs LuckyLobster) and
|
||||
what strategies still work
|
||||
- Analyze why Anthropic is dominating the "best model" market while OpenAI leads in
|
||||
IPO hype
|
||||
|
||||
✻ Cogitated for 4m 18s
|
||||
```
|
||||
Reference in New Issue
Block a user