Skip to main content
FlowForge
How It Works

Four steps from raw data to intelligence

Observe → Analyse → Report → Monitor. Every step is transparent, every output is evidence-based, every observation carries full provenance.

STEP 1

Observe

Nine independent engines ingest on-chain data, developer activity, market structure, and smart money flow in real-time. Each engine tags its output with source mode (LIVE/MOCK/DEGRADED/UNKNOWN).

STEP 2

Analyse

The evidence fusion engine resolves all engine outputs into a single intelligence state. Conflicting evidence is surfaced honestly — we show CONFLICTED rather than pretending clarity that doesn't exist.

STEP 3

Report

Constitutional review enforces honesty rules: UNKNOWN over fabrication, no BUY/SELL language, transparent source modes. Every observation carries full provenance — which engines, what evidence, what confidence.

STEP 4

Monitor

Observations update as evidence changes. Your dashboard shows real-time intelligence states, evidence tabs per token, and wallet-level observations — all in observation mode, never trade signals.

Evidence Hierarchy

Every data point in the pipeline is tagged with its source mode. This is always visible — MOCK is never hidden.

LIVEData from real providers. Production-grade intelligence.
MOCKSynthetic data for testing. Never disguised as live.
DEGRADEDPartially live. Some engines on fallback data.
UNKNOWNSource cannot be determined. Treated as unreliable.

Confidence Hierarchy

██████████LIVE — Full confidence, production data
████████░░DEGRADED — Partial, some engines on fallback
█████░░░░░MOCK — Synthetic, testing only
██░░░░░░░░UNKNOWN — Cannot determine source reliability

Constitutional rule: UNKNOWN is always preferable to fabrication. If we don't have data, we say so.

What We Don't Do

Understanding what FlowForge is not is just as important as understanding what it is.

✗ We Don't

  • • Give buy/sell recommendations
  • • Provide entry/exit zones or price targets
  • • Claim win rates as validated outcomes
  • • Disguise MOCK data as LIVE
  • • Promise returns or profit
  • • Execute transactions on your behalf

✓ We Do

  • • Provide observational intelligence states
  • • Show evidence with full engine provenance
  • • Label source mode on every data point
  • • Surface conflicting evidence honestly
  • • Analyse your wallet (read-only, no keys)
  • • Give you intelligence to make your own decisions

Start Observing

Enter the V2 Lab to see real-time token observations with full evidence provenance.

Start ObservingRead the Method

Not financial advice. FlowForge provides observational intelligence for educational and informational purposes only. It does not constitute financial advice, investment advice, or trading recommendations. Digital assets involve substantial risk of loss. You remain solely responsible for your own decisions.

Past outcomes do not guarantee future results. Historical observations, accuracy rates, and performance metrics are backward-looking and should not be interpreted as a guarantee or prediction of future performance.

Evidence confidence, not probability. Confidence scores measure the strength, coverage, freshness, and consistency of available evidence. They are not automatically a probability of profit or future market movement. A high confidence score means more supporting evidence — it does not mean a trade will be profitable.

Data source modes: All data is tagged with its provenance — LIVE (real-time production data), HYBRID (mixed live and sample), SAMPLE (representative sample for development), or UNKNOWN (source cannot be confirmed). Sample and hybrid data must never be treated as live intelligence.

Your responsibility. You should conduct your own independent research, consider your financial situation and risk tolerance, and never invest more than you can afford to lose. Consult a qualified financial advisor if needed.

MethodologyRisk DisclosurePrivacyTerms