Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Synthesize all per-scenario evaluations into a final portfolio recommendation with robustness score and actionable guidance.
.claude/skills/yogsoth-ai-portfolio-synthesis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 115% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 165% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 50% | 0% |
Aggregate evaluations across all scenarios into a final portfolio recommendation, robustness assessment, and actionable guidance.
Spawns a subagent that synthesizes per-scenario results into a coherent overall assessment with clear recommendation.
Synthesis requires integrating multiple scenario evaluations, identifying patterns across scenarios, and making a holistic judgment. This integrative reasoning benefits from seeing all data together with focused attention.
Output must include a final portfolio recommendation, numeric robustness score (0-1), and specific actionable recommendations. The synthesis must reference findings from all evaluated scenarios.
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | spawn-agent | Spawn a customized CC subagent with full MCP tool access. Used by SOPs that declare execution: subagent. |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 9,887 | 10,079 | +2% | 1 | 1 | 0% | 1,117 | 1,204 | +8% | 0 | 0 | — |
case-01 | fail→pass | 15,875 | 36,480 | +130% | 1 | 1 | 0% | 2,732 | 5,870 | +115% | 0 | 0 | — |
case-02 | fail→pass | 20,693 | 22,149 | +7% | 1 | 1 | 0% | 2,639 | 2,910 | +10% | 0 | 0 | — |
case-03 | fail→pass | 28,643 | 24,448 | -15% | 1 | 1 | 0% | 3,414 | 4,162 | +22% | 0 | 0 | — |
case-04 | fail→pass | 19,039 | 51,818 | +172% | 1 | 1 | 0% | 2,882 | 7,623 | +165% | 0 | 0 | — |
case-05 | fail→pass | 20,930 | 28,774 | +37% | 1 | 1 | 0% | 2,575 | 3,858 | +50% | 0 | 0 | — |
case-06 | fail→pass | 27,138 | 33,328 | +23% | 1 | 1 | 0% | 3,336 | 4,725 | +42% | 0 | 0 | — |
case-07 | fail→pass | 25,062 | 36,841 | +47% | 1 | 1 | 0% | 3,002 | 5,118 | +70% | 0 | 0 | — |
case-08 | fail→pass | 18,823 | 38,984 | +107% | 1 | 1 | 0% | 2,100 | 5,659 | +169% | 0 | 0 | — |
case-09 | fail→pass | 24,236 | 29,627 | +22% | 1 | 1 | 0% | 3,574 | 3,893 | +9% | 0 | 0 | — |
case-10 | fail→pass | 13,147 | 24,819 | +89% | 1 | 1 | 0% | 1,239 | 3,196 | +158% | 0 | 0 | — |
case-11 | fail→fail | 23,273 | 12,318 | -47% | 1 | 1 | 0% | 2,621 | 447 | -83% | 0 | 0 | — |
case-12 | fail→pass | 17,512 | 26,393 | +51% | 1 | 1 | 0% | 1,749 | 3,538 | +102% | 0 | 0 | — |
case-13 | fail→pass | 14,460 | 36,414 | +152% | 1 | 1 | 0% | 2,108 | 3,567 | +69% | 0 | 0 | — |
case-14 | fail→pass | 17,175 | 32,731 | +91% | 1 | 1 | 0% | 1,713 | 4,500 | +163% | 0 | 0 | — |
case-15 | fail→pass | 14,954 | 27,711 | +85% | 1 | 1 | 0% | 2,205 | 3,816 | +73% | 0 | 0 | — |
case-16 | fail→fail | 19,212 | 12,974 | -32% | 1 | 1 | 0% | 2,147 | 742 | -65% | 0 | 0 | — |
case-17 | fail→pass | 17,529 | 36,501 | +108% | 1 | 1 | 0% | 1,819 | 3,245 | +78% | 0 | 0 | — |
case-18 | fail→pass | 21,561 | 33,365 | +55% | 1 | 1 | 0% | 2,360 | 4,742 | +101% | 0 | 0 | — |
case-19 | fail→fail | 52,190 | 38,129 | -27% | 1 | 1 | 0% | 8,230 | 5,597 | -32% | 0 | 0 | — |
case-20 | pass→pass | 12,413 | 12,854 | +4% | 1 | 1 | 0% | 1,442 | 1,663 | +15% | 0 | 0 | — |
case-22 | pass→pass | 27,680 | 32,386 | +17% | 1 | 1 | 0% | 3,028 | 6,540 | +116% | 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. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 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 +73 percentage points is the difference between those two pass rates over the 20 comparable cases.
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.