⌬ Multi-Agent Council

Strategy Forge

15-agent pipeline: propose → code → backtest → critique (12 specialists in parallel) → decide. Every run either GRADUATEs (rare), REFINEs (most common), PIVOTs (rare), or ABANDONs.

What this page is telling you

Loading…

Click a run row below to drill in. Look for runs marked GRADUATE — those are the strategies the council says are ready to deploy. ABANDON verdicts are also instructive — they tell you what doesn't work.

Total Runs
runs + EA reviews through the council
Graduated
consensus: ship
Abandoned
killed by the council
In Progress
verdict pending
How it works propose → code → backtest → critique → decide

A strategy proposer drafts a hypothesis. The code-writer agent translates it into a tested Python module. The backtester runs it on real historical data. Then twelve specialist critics — each looking at a different failure mode — independently vote on whether to graduate, refine, pivot, or abandon. The arbitrator synthesizes their verdicts into a single decision and writes the next round's brief.

Statistician
Risk-Engineer
Quant-Skeptic
Regime Analyst
Execution Realist
Microstructure
Trend-Trader
Mean-Reversion
Macro-Strategist
Discretionary
Systematic-Quant
Prop-Firm Vet
Verdict Outcomes
GRADUATECouncil consensus says ship
REFINEIterate on hypothesis
PIVOTRestart with new thesis
ABANDONNo path to viability

Historically < 5% of runs graduate on round 1. Most are REFINEs that go 3–5 rounds before reaching a final verdict. ABANDONs are decisive: the council saw a fundamental flaw that no parameter tweak will fix.

Strategy Runs

loading…
Loading forge runs…