Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Optimize portfolio allocation using npx neural-trader mean-variance engine with risk constraints and rebalancing plan
.claude/skills/ruvnet-trader-portfolio/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -6% | 0% |
Optimize portfolio allocation using neural-trader's portfolio engine.
Steps:
npm ls neural-trader 2>/dev/null || npm install --ignore-scripts neural-trader
mcp__plugin_ruflo-core_ruflo__memory_search({ query: "current portfolio holdings", namespace: "trading-portfolio" })
bash npx neural-trader --portfolio optimize With risk target: bash npx neural-trader --portfolio optimize --risk-target <number>
bash npx neural-trader --risk assess --portfolio current npx neural-trader --var --portfolio current npx neural-trader --correlation --portfolio current --flag-threshold 0.8
mcp__plugin_ruflo-core_ruflo__neural_predict({ input: "expected returns for [HOLDINGS] given current regime" })
bash npx neural-trader --portfolio rebalance Output: trades needed, current vs target weights, estimated costs
mcp__plugin_ruflo-core_ruflo__agentdb_pattern-search({ query: "optimized portfolio Sharpe > 1", namespace: "trading-portfolio" })
mcp__plugin_ruflo-core_ruflo__memory_store({ key: "portfolio-optimal-TIMESTAMP", value: "ALLOCATION_JSON", namespace: "trading-portfolio" })
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 9,153 | 6,242 | -32% | 1 | 1 | 0% | 1,839 | 1,694 | -8% | 0 | 0 | — |
case-01 | fail→fail | 14,925 | 18,712 | +25% | 1 | 1 | 0% | 3,060 | 971 | -68% | 0 | 0 | — |
case-02 | fail→fail | 13,765 | 7,746 | -44% | 1 | 1 | 0% | 2,512 | 1,022 | -59% | 0 | 0 | — |
case-03 | fail→fail | 21,284 | 6,073 | -71% | 1 | 1 | 0% | 4,422 | 855 | -81% | 0 | 0 | — |
case-04 | fail→fail | 36,970 | 3,633 | -90% | 1 | 1 | 0% | 3,922 | 1,060 | -73% | 0 | 0 | — |
case-05 | fail→pass | 9,271 | 2,738 | -70% | 1 | 1 | 0% | 1,730 | 840 | -51% | 0 | 0 | — |
case-22 | pass→pass | 11,880 | 17,569 | +48% | 1 | 1 | 0% | 2,402 | 2,337 | -3% | 0 | 0 | — |
case-06 | fail→fail | 8,695 | 3,038 | -65% | 1 | 1 | 0% | 1,521 | 954 | -37% | 0 | 0 | — |
case-07 | fail→pass | 15,745 | 9,386 | -40% | 1 | 1 | 0% | 924 | 1,446 | +56% | 0 | 0 | — |
case-08 | fail→fail | 9,337 | 7,170 | -23% | 1 | 1 | 0% | 1,663 | 1,089 | -35% | 0 | 0 | — |
case-09 | fail→pass | 8,967 | 3,641 | -59% | 1 | 1 | 0% | 1,392 | 1,123 | -19% | 0 | 0 | — |
case-10 | fail→fail | 10,596 | 4,295 | -59% | 1 | 1 | 0% | 1,848 | 689 | -63% | 0 | 0 | — |
case-11 | fail→pass | 9,169 | 2,172 | -76% | 1 | 1 | 0% | 1,573 | 805 | -49% | 0 | 0 | — |
case-12 | fail→fail | 12,629 | 3,902 | -69% | 1 | 1 | 0% | 2,459 | 1,111 | -55% | 0 | 0 | — |
case-13 | fail→fail | 10,893 | 7,053 | -35% | 1 | 1 | 0% | 2,148 | 1,084 | -50% | 0 | 0 | — |
case-14 | fail→pass | 5,375 | 5,876 | +9% | 1 | 1 | 0% | 1,022 | 956 | -6% | 0 | 0 | — |
case-15 | fail→pass | 9,835 | 5,120 | -48% | 1 | 1 | 0% | 2,008 | 1,458 | -27% | 0 | 0 | — |
case-16 | fail→pass | 11,524 | 2,769 | -76% | 1 | 1 | 0% | 2,142 | 969 | -55% | 0 | 0 | — |
case-17 | fail→pass | 10,395 | 3,105 | -70% | 1 | 1 | 0% | 1,642 | 974 | -41% | 0 | 0 | — |
case-18 | fail→pass | 14,071 | 4,676 | -67% | 1 | 1 | 0% | 3,017 | 1,412 | -53% | 0 | 0 | — |
case-19 | fail→pass | 9,939 | 7,031 | -29% | 1 | 1 | 0% | 1,784 | 1,803 | +1% | 0 | 0 | — |
case-20 | fail→fail | 13,158 | 6,383 | -51% | 1 | 1 | 0% | 2,369 | 1,773 | -25% | 0 | 0 | — |
case-23 | pass→pass | 18,242 | 22,741 | +25% | 1 | 1 | 0% | 4,232 | 5,687 | +34% | 0 | 0 | — |
case-24 | pass→fail | 20,717 | 4,869 | -76% | 1 | 1 | 0% | 4,178 | 702 | -83% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted, and 16 counted toward the lift figure. The other 8 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +38 percentage points is the difference between those two pass rates over the 16 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.