Copytrading powered by an onchain intelligence protocol.Live onchain

The edge should
spill over.

Copy a public wallet instantly. Choose regular, agent, or split mode. Both paths buy immediately; the agent researches afterward, then holds or exits. Every consenting result strengthens the shared global brain.

NO ALLOWLIST AT LAUNCH. ANY CONSENTING SPILL HOLDER CAN SET CAPPED LIVE-TRADING PERMISSIONS.

SAMPLE COMPARISON / 024 EXAMPLE ONLY
REGULAR COPY+2.4%
AGENT COPY+4.1%
DIFFERENCE+1.7pp

Sample data only. These numbers are not live results or a performance claim.

00 / HOW IT WORKS

One whale buy.
Two immediate paths.

Choose regular, agent, or split mode. Every selected path buys immediately. Only the agent researches afterward.

PUBLIC WALLET WHALE BUYS ROBINHOOD CHAIN
REGULAR COPYBENCHMARK
BUY
HOLD
AGENT COPYGLOBAL BRAIN
BUY
RESEARCH
+ REASON
AFTER ENTRY, DECIDE
LOOKS GOOD KEEP
TOO RISKY CLOSE ASAP

Paper agent. Simulates the buy and decision. No second spend.

Real agent. Uses only its assigned share. A $10 budget at 50/50 becomes $5 regular + $5 agent.

Risk looks high? Exit as soon as the entry confirms.

01

Opt in. Learning starts only after signed consent. Revoke anytime.

02

Every result counts. Holds, exits, wins, and losses become evidence.

03

Changes must earn promotion. New policies stay in shadow until they beat regular copy on unseen trades.

01 / GLOBAL BRAIN

Nobody starts
from zero.

CUMULATIVE BY DESIGN

With consent, every outcome joins one shared learning loop. Validated lessons improve the policy used by future agent trades.

HOW THE NETWORK LEARNS

One learning loop, shared by every holder.

  1. 01 / SIGNALOBSERVE

    Research once.
    Share everywhere.

    Each source trade is researched once. Every eligible follower shares the result.

  2. 02 / BASELINECOMPARE

    Compare the
    same entry.

    Regular and agent paths start together. Costs and exits stay in the comparison.

  3. 03 / MEMORYLEARN

    Turn outcomes
    into memory.

    Results mature at 5m, 30m, 4h, and 24h. Holds, exits, wins, and losses all count.

  4. 04 / POLICYPROMOTE

    Prove it before
    promotion.

    New policies must win on unseen trades without adding drawdown.

THE COMPOUNDING UTILITY

The network learns once.
Every holder gets the update.

More use creates more evidence. Better evidence improves one policy shared by everyone.

05 / TEACH THE BRAIN

Bring an insight.
Prove the edge.

Submit a public source and your read. It enters shadow testing and cannot change live policy alone.

AGENT EDGE

Beat regular copy. Earn from the pool.

When your real agent leg wins after costs, its measured edge earns points.

KNOWLEDGE EDGE

Improve the brain. Earn from the pool.

Evidence that survives review and improves unseen trades earns points.

10% of collected SPILL pool fees funds the Edge Pool. Reviewed agent edge and useful evidence earn points. After each seven-day challenge window, eligible users claim ETH or SPILL from the onchain escrow. No mainnet epoch exists yet.

LIVE NETWORK LEDGERSYNCING
SOURCE TRADES ANALYZED
PAIRED SHADOW RUNS
MATURED 24H OUTCOMES
PROMOTION GATE / 200
02 / ADAPTIVE INTELLIGENCE

The flywheel
buys more brain.

40% of pool fees funds research. The agent starts with onchain checks, then unlocks Nansen, Grok 4.5 search, and higher compute only when revenue can support it.

CURRENT RESEARCH MACHINEONCHAIN CORE
RESEARCH CASH AVAILABLE$0
FORECAST UNIQUE TRADES / MONTH1,000

Waiting for settled fee revenue or a reviewed bootstrap credit.

  1. 00 / ALWAYS ON

    Onchain Core

    Contract, market, explorer, and shared-policy checks.

    $0 / TRADECURRENT
  2. 01 / SCOUT

    Grok 4.5 · Low

    Nansen profile + fast live web search.

    $0.03 RESERVED / TRADE · 100K STAKE CEILINGCALCULATING
  3. 02 / RESEARCH

    Grok 4.5 · Medium

    More reasoning + web and X search with sources.

    $0.09 RESERVED / TRADE · 500K STAKE CEILINGCALCULATING
  4. 03 / DEEP

    Grok 4.5 · High

    Deeper web and X investigation.

    $0.20 RESERVED / TRADE · 2.5M STAKE CEILINGCALCULATING
  5. 04 / SURPLUS

    GPT‑5.6 Sol · High

    Frontier research after a three-month runway is covered.

    $1.00 RESERVED / TRADE · 10M STAKE CEILINGCALCULATING

SUSTAINABILITY RULE One analysis per source trade. Followers share the result. Staking can raise the compute ceiling, but never overrides the operations floor; the system falls back when cash or a provider is unavailable.

03 / STAKING BENEFITS

Stake SPILL.
Unlock more brain.

TESTNET VERIFIED · MAINNET PENDING

The reward is more product power—not token yield. Everyone keeps the fee-funded global baseline. Active stake raises the maximum research machine the agent can use and the number of public wallets you can follow.

WITHOUT STAKINGGlobal agent baseline

Follow up to 10 source wallets.

WITH ACTIVE STAKEDeeper research ceiling

Follow up to 100 source wallets.

Stake in Copy Lab
MODEL POLICY / ALWAYS CURRENT

Your stake unlocks a capability class, not one permanent model. The engines below are the current lineup. New releases are evaluated for research quality, speed, and sustainable cost, then replace an incumbent when they are a better fit. Your tier carries forward to the upgraded model.

  1. 01 / SCOUT 100K SPILL LITE COMPUTE

    Grok 4.5 · Low

    Fast web and X research after entry.

    20 SOURCE WALLETS
  2. 02 / RESEARCH 500K SPILL FULL RESEARCH

    Grok 4.5 · Medium

    More search depth and reasoning turns.

    35 SOURCE WALLETS
  3. 03 / DEEP 2.5M SPILL DEEP REASONING

    Grok 4.5 · High

    Deeper live investigation before hold or exit.

    60 SOURCE WALLETS
  4. 04 / SURPLUS 10M SPILL TOP FRONTIER

    GPT‑5.6 Sol · High

    Frontier reasoning when the compute pool can fund it.

    100 SOURCE WALLETS
7 DAYS

Warm-up before benefits activate.

IMMEDIATE

Benefits end when unstaking starts.

7 DAYS

Cooldown before principal withdrawal.

NEVER

No APY, slashing, or protocol access to principal.

FEE-BUDGETED BY DESIGN A stake sets your ceiling, not a guaranteed spend. The 40% compute pool must be able to pay for the stronger research tier; otherwise the agent uses the best funded tier without delaying the copied entry.

04 / COPY LAB

Copy the trade.
Make the agent earn it.

The preview below is visual only. Open the Copy Lab to connect a wallet, follow a source, and set real guardrails.

COPY LAB

Copy console

Robinhood Chain · 4663
SOURCE WALLET
7A
0x7A19…F42CSample wallet · preview data
82SAMPLE
30D RETURN+18.6%
WIN RATE61%
MAX DD−7.2%
TRADES43
SAMPLE COMPARISON

Regular copy vs agent copy

SHADOW
PATHVALUERETURNMAX DDSTATE
Regular copyBenchmark $1,024.20+2.42%−4.8%RUNNING
Agent runShared policy v0 $1,041.10+4.11%−3.1%+1.69PP
×
SAMPLE AGENT DECISION · 14:32:08Closed MICA after a contract risk check

Both paths entered immediately. Post-entry research found elevated contract risk, so the agent exited while the base benchmark stayed open.

84%CONFIDENCE
YOUR GUARDRAILS
Risk checks on Direct Uniswap swaps Starts from next trade

Illustrative interface. No wallet or API is connected on this page.

Homepage preview. Open the Copy Lab to connect an execution wallet and start a real monitor. Open copy lab
05 / FEE ENGINE

Trading fees
pay for the product.

SPILL uses a 1% Uniswap v4 pool fee. At launch, opening buys begin at 80% and fall to 1% over 10 seconds; sells stay at 1%. Fees split automatically: 40% compute, 10% Edge Pool, 30% founder, 20% reserve. Verified on Robinhood testnet. Mainnet is not deployed.

1.00%NORMAL POOL FEE
BEFORE PROTOCOL FEE
40%COMPUTE + OPS

Agent research, RPC, monitoring, and cloud runtime.

10%EDGE POOL

Reviewed agent-edge and evidence payouts.

30%FOUNDER

Fixed creator revenue.

20%PROTOCOL RESERVE

Audits, incidents, and liquidity support.

VOLUME MODEL

What does the flywheel produce?

MONTHLY SPILL VOLUME$1.00M
SPILL LP FEES$10,000
COMPUTE$4,000
EDGE POOL$1,000
FOUNDER$3,000
RESERVE$2,000

Steady-state 1% illustration. Excludes launch protection, protocol fees, costs, taxes, and price changes.

THE PREMISE
Whales leave a wake.
We build with what spills over.

Whale trades are public. Spillover copies them, learns from every consenting outcome, and shares each improvement with every holder. Token volume funds the loop.

Start live copying