NOT READY. Three data-layer CRITICALs allegedly block any real-money sign-off:
asset_class_health.n=0 (now known to be read-error; see correction above)Holding survivors of the cross-check: drift alert TRUE but KS_D uncomputed; baby_strats 12-flag overfit; system draggers (kimi_signal_tracking -954%, mercury2_fast, ml_bg_system_b); claude_gainer_st winner-vs-blacklist contradiction (in BLACKLISTED_STRATEGIES at alpha_engine/config.py:216 but tops systems with PF 6.12).
| Rank | System | PF | WR % | n | MDD % | Flag |
|---|---|---|---|---|---|---|
| 1 | multi_asset_cot | 19.19 | 86.9 | 130 | 17.8 | implausible PF — verify or quarantine |
| 2 | aggregated_picks | 6.42 | 77.3 | 441 | 14.4 | aggregator artifact suspect |
| 3 | claude_gainer_st | 6.12 | 78.5 | 3472 | 14.0 | in BLACKLISTED — contradiction |
| 4 | ml_crypto_pred_v12 | 2.53 | 55.6 | 123 | 11.0 | likely DEAD (no last_signal) |
| 5 | signal_validation | 2.22 | 55.0 | 502 | 14.1 | validator-as-pick-source meta-eval risk |
| 6 | copy_trader_intel | 1.84 | 50.0 | 739 | 2.23 | MDD 2.23% looks synthetic |
All 6 lack last_signal_date. Treat as DEAD until proven live.
| System | PnL % | PF |
|---|---|---|
| kimi_signal_tracking | -954.9 | 0.26 |
| multi_asset | -146.0 | 0.31 |
| mercury2_fast | -139.5 | 0.07 |
| alpha_engine_fast | -127.6 | 0.62 |
| paper_trading | -103.5 | 0.35 |
| copy_trader_highscore | -81.8 | 0.76 |
| ml_bg_system_b | -54.7 | 0.02 |
| crypto_winners | -52.4 | 0.36 |
| Priority | Action | Effort | Risk |
|---|---|---|---|
| P0 | Verify multi_asset_cot PF=19.19 via DB query against ejaguiar1_stocks | 1h | Low |
| P0 | Reconcile claude_gainer_st winner-vs-blacklist contradiction; confirm enforcement at exec gate (memory feedback_gate_at_execution_not_generation) | 1-2h | Low |
| P0 | Refresh stale dashboard_data.json; diagnose refresh-cron failure via check-gh-actions | 0.5h | Low |
| P1 | Execute existing baby_strats 12-flag quarantine proposal | 1h | Low |
| P1 | Fix drift detector — KS_D uncomputed-zero + hf_stats 19d stale | 3-4h | Med |
| P1 | Investigation-gate top draggers per MUTATION_THREE_AXIS_PROTOCOL.md | 4-6h | Med |
| P1 | Add last_signal_date to systems payload | 1-2h | Low |
| P2 | Audit HC filter against confidence-inversion on ETF/CRYPTO (memory project_performance_reality) | 2-3h | Low |
| P2 | Re-run cross-permutations; verdict on per-class edge existence | 1h | Low |
Ranked by impact/effort: Riskfolio-Lib (CVaR/HRP), QuantStats (reporting), pandas-ta (130+ indicators), FRED (macro filter — fred_data_fetcher.py exists, needs key), VectorBT (50-100× backtests), Kalshi (Polymarket pairwise), PyPortfolioOpt, Glassnode/Coinglass widen, Freqtrade, FinRL.
All subject to CLAUDE.md Wire-Up Rule — production caller required or explicit opt-in sidecar with Wiring Plan section.
Plan file persisted at C:\Users\zerou\.claude\plans\follow-that-skill-and-joyful-lemon.md (developer workstation, not committed). Contains: