Snapshots logged 3
Last snapshot
On-chain signals 5
Contract 0xFFCC…F453
Agent registered
ERC-8004 ID 1470
Network BSC Testnet

When markets close,
bStocks keep drifting.
BQuant measures the gap.

NYSE and Nasdaq halt every weekend. The bStock tokens that mirror them on BNB Chain do not. Over those closed hours, on-chain price and reference price diverge — and when markets reopen Monday, gravity pulls them back. BQuant tracks that drift in real time and emits a mean-reversion signal the moment the gap is wide enough to act on.

Tokens  NVDAB · TSLAB · CRCLB · MUB · SNDKB
Chain  BNB Chain (BEP-20)
Data source  CMC Skill Hub MCP
Signal engine  drift_engine.py

Current drift readings

Live data, updated automatically — not backfilled or simulated.

NVDAB · NVIDIA −0.0353% Converged $209.4860 anchor $209.56 · conf 90
TSLAB · Tesla +0.0805% Converged $400.8524 anchor $400.53 · conf 90
CRCLB · Circle +0.0514% Converged $80.6615 anchor $80.62 · conf 90
MUB · Micron +0.2552% Converged $1,134.8839 anchor $1,131.995 · conf 90
SNDKB · SanDisk +0.1168% Converged $2,243.2475 anchor $2,240.63 · conf 90

Drift % from anchor · all 5 tokens · 3 snapshots (22:46 → 23:15 UTC, June 20 2026)

On-chain signal log (BSC Testnet)

5 signals logged to BQuantSignalRegistry at 0xFFCC472c47cf0a8168545a8318832950f7C6F453. On-chain enum: FLAT=0, LONG=1, SHORT=2. CONVERGED maps to FLAT for gas efficiency.

logId Token Signal (on-chain) Drift (bps) Confidence Transaction
0 NVDAB FLAT / CONVERGED −4 bps 90 0x5956…e0a ↗
1 TSLAB FLAT / CONVERGED +8 bps 90 0x14a2…f7f ↗
2 CRCLB FLAT / CONVERGED +5 bps 90 0x784f…dda ↗
3 MUB FLAT / CONVERGED +26 bps 90 0x491f…2df ↗
4 SNDKB FLAT / CONVERGED +12 bps 90 0xf9b8…038 ↗

The weekend gap

bStocks trade around the clock — but the real stocks they track (NVIDIA, Tesla, Micron…) only trade during NYSE and Nasdaq hours. When the real market is closed, there's nothing keeping the on-chain price in line. It can drift.

BQuant checks prices every 30 minutes and compares each one to the starting price (the anchor). If a token has drifted far enough, the bet is that it snaps back once the real market reopens. So BQuant signals the opposite of the drift — price went up means SHORT (expect it to fall back); price went down means LONG (expect it to rise back). Small drift means no signal yet.

What happened Signal Logic Confidence
Price moved more than 1.5% LONG / SHORT Signal opposite to direction 60 → 85 (scales with size)
Price is within 0.3% of anchor CONVERGED Already back where it should be Fixed 90
Anywhere in between FLAT Not enough signal yet Fixed 50

Every signal includes this block

[Strategy: BQuant — bStock Weekend Drift]
[Token: <TICKER>]
[Anchor (t0): <price> @ <timestamp>]
[Current Price: <price> @ <timestamp>]
[Drift: x.xx%]
[Signal: LONG / SHORT / FLAT / CONVERGED]
[Confidence: xx%]
[Reasoning: <one sentence>]

Add it to your agent

Drop BQuant's signal engine into any agent.

from drift_engine import calculate_drift, generate_signal

drift = calculate_drift(snapshots, "NVDAB")
signal = generate_signal(drift["drift_pct"])
# → {"signal": "LONG", "confidence": 85, "reasoning": "..."}

That's it — pass in your own price snapshots, get a signal back. Full reference in the strategy doc →