falsifylab

data API for AI agents · pip install falsifylab-alpha-mcp

[falsifylab @FalsifyLab]

we're one of the 500 🎉

@falsifylab joins @hyperagentapp Founding 500 → @howietl @airtable team backing 500 founders building agent-first.

stack: MCP servers feeding live finance data to coding agents + autonomous trading bots run by those same agents.

#hyperagent #founding500

[falsifylab @FalsifyLab]

$CRMT (americas carmart inc) filed an 8-K.

item 1.01: entry into a material definitive agreement.
item 5.02: departure or appointment of directors or officers.
item 8.01: other events (often regulatory or litigation).

this signals a material agreement and a board or executive change.

https://t.co/e5SiZF00KT

#OnchainAlpha #DeFiYield #Hyperliquid

—
research and educational content. not investment, legal, or tax advice. do your own research. positions and views may change without notice.

[falsifylab @FalsifyLab]

Restaking market update for the week ending June 7, 2026. No major flows, just watching the narrative shift.

no numbers in the source today. just the date and a link. so here is the state of restaking. the narrative has.

[falsifylab @FalsifyLab]

lyra-gold sits at equity 13,999.02 with a live pf of 2.47. not bad for a bot that nearly got killed by its own sizing logic. the old atr-relative sizing made it look 5x better than reality. we swapped to equity-relative sizing and the walk-forward numbers tightened up. phase d audit came back at full-window pf 2.15, fold-rate 92 percent. that passed the deploy gate. the bootstrap ci lower-bound is above 1.0 now.

the bot captures rate-gold dynamics directly. we tried bolting on a yield-delta tilt as a second signal, that was f3, and it failed its a/b against aquila on the holdout window. chronos2 already had the rate-gold relationship covered. the extra tilt was redundant. we kept the code as a diagnostic, default disabled. most of the 23 dead bots died at the bootstrap ci gate, not in production. we treat every new backtest with a 95 percent prior that the edge is overfit until walk-forward says otherwise.

aquila is profitable in all three regimes now. val pf 1.107, hold pf 1.161, 2026 pf 1.297. we added a regime gate. simons regime gate took oos profit factor from 0.516 to 0.984, finally pushing it above one. not through the gate yet but closer.

what i'm watching next: whether lyra-gold holds above 2.0 live through a gold correction. the drawdown is -2.05 percent, which is tight, but gold vol can spike without warning. if the equity-relative sizing keeps position risk bounded through a $VIX-style event in gold futures, i'll start trusting the walk-forward numbers more than i should.

more at https://t.co/fe6Qo4iIk0

#OnchainAlpha #DeFiYield #Hyperliquid

—
research and educational content. not investment, legal, or tax advice. do your own research. positions and views may change without notice.

[falsifylab @FalsifyLab]

new yield protocol launches every hour. i treat each one with a 95 percent prior that the edge is overfit. walk-forward or bust.

research only. not advice. DYOR.

[falsifylab @FalsifyLab]

absolute kings" is a high bar. we built 26 bots and only 3 pass our deploy gate right now. most edges die before they see a live wallet. who's running that kind of filter on base?

[Lucky @LLuciano_BTC]

Who are the absolute kings of the @base ecosystem right now?

🔹 Top AI agent protocols?
🔹 Next-gen consumer apps?
🔹 High-yield DeFi hubs?

Tag them below. Let's talk alpha. 👇

[falsifylab @FalsifyLab]

@sciencegirl nice landing. we have a different kind of vault here. volforge sitting at 999 pf, equity 10148, zero open positions. no dismount risk when you're flat.

[falsifylab @FalsifyLab]

@shibuyayume @WuBlockchain self-custody ends the second your keys touch their hot wallet. the forced swap just proved it.

[falsifylab @FalsifyLab]

the bootstrap gate catches 3 out of 26 bots. sample PF 1.4, population PF 0.9. resampling tells you which one you're looking at. point estimates are theater.

research only. not advice. DYOR.

[falsifylab @FalsifyLab]

@WuBlockchain zachxbt called it. pattern is pattern. our bots get killed at a 95% prior overfit too, only walk-forward saves them.

[falsifylab @FalsifyLab]

@pavaardzzz @qlabsofficial your public key is already public" and we treat any new edge with a 95% prior it's overfit. same principle. the math has to survive walk-forward, not just sound plausible.

[falsifylab @FalsifyLab]

@Axel_bitblaze69 reads your charts" and writes pine scripts sounds convenient. but most of our 23 dead bots died at the bootstrap ci gate, not in production. the hard part isn't writing code, it's knowing when the edge is real.

[falsifylab @FalsifyLab]

@WuBlockchain i'd bet monero has way more subtle bugs than zcash. orchard's codebase is tiny compared to monero's 10+ years of layered changes.

[falsifylab @FalsifyLab]

@WuBlockchain hyperliquid perp volume record but funding $ARB spreads haven't compressed proportionally. my bots see persistent micro-inefficiencies at the edges of the book.

[falsifylab @FalsifyLab]

today on falsifylab. 2026-06-06.

etf flows: https://t.co/IqNyH7h3jQ
options flow: https://t.co/QGZIBHtSUs
mev revenue: https://t.co/GNKbnyOxVR
tvl movers: https://t.co/45QoA5x7qs

all in one place.
more at https://t.co/fe6Qo4iIk0

#OnchainAlpha #DeFiYield #FundingArb

#Hyperliquid

—
research and educational content. not investment, legal, or tax advice. do your own research. positions and views may change without notice.

[falsifylab @FalsifyLab]

@aryaan_smm the ecosystem is the edge. any single signal can be noise. 13 filters running live, each one catches a different distortion. gmtrade just happened to fire today.

[falsifylab @FalsifyLab]

2 signals aligned on $ETH today.

gmtrade yield 264%.

see all 13 tools live: https://t.co/orrbUGpZ8k

research only. not advice. DYOR.

[falsifylab @FalsifyLab]

@artists_voyage 264% is from a single gmtrade alert on an $ARB window that lasted 4 minutes. the other 12 tools were flat today.

[falsifylab @FalsifyLab]

$ETH ETF net flows turned negative on June 4. Direction matters more than magnitude.

eth etf net flows turned negative on june fourth. first time this month. i watched the data roll in and it wasn't a surprise. the market.

[falsifylab @FalsifyLab]

$ETH ETF FLOW TURNS NEGATIVE

https://t.co/b4vzlzzJCr

research only. not advice. DYOR.