Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Synthesize constraint analysis into actionable report with priorities
.claude/skills/yogsoth-ai-constraint-synthesis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -41% | 0% |
Synthesize all constraint analysis outputs into a single actionable report. Integrates findings from tree-building, sensitivity ranking, and constraint-breaking into prioritized recommendations.
Subagent — spawned via subagent-spawning/spawn-agent skill.
<!-- BEGIN available-tables (generated) --> <!-- context-management rows hand-maintained (cross-package import); do not regenerate this file -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | context-checkpoint | Append research process and results to the current Phase's context file. Each append MUST contain >=500 lines of markdown covering both process and results. Use this skill at plan-designated checkpoint points — typically after each strategy completes or at key decision nodes within a research Phase. | | context-init | Create a new context file for a research Phase. Called once at Phase start to initialize the file that subsequent context-checkpoint calls will append to. Use this skill whenever a new research Phase begins and a fresh context file is needed. | | 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 | 7,683 | 1,977 | -74% | 1 | 1 | 0% | 1,117 | 602 | -46% | 0 | 0 | — |
case-01 | fail→fail | 16,185 | 23,198 | +43% | 1 | 1 | 0% | 2,458 | 3,812 | +55% | 0 | 0 | — |
case-02 | fail→fail | 13,517 | 14,097 | +4% | 1 | 1 | 0% | 2,020 | 2,386 | +18% | 0 | 0 | — |
case-03 | fail→pass | 16,509 | 19,273 | +17% | 1 | 1 | 0% | 2,308 | 3,227 | +40% | 0 | 0 | — |
case-04 | fail→pass | 3,830 | 2,769 | -28% | 1 | 1 | 0% | 594 | 749 | +26% | 0 | 0 | — |
case-05 | fail→pass | 4,373 | 3,561 | -19% | 1 | 1 | 0% | 718 | 814 | +13% | 0 | 0 | — |
case-06 | pass→pass | 15,128 | 20,052 | +33% | 1 | 1 | 0% | 2,452 | 3,567 | +45% | 0 | 0 | — |
case-07 | pass→pass | 10,181 | 14,593 | +43% | 1 | 1 | 0% | 2,124 | 3,392 | +60% | 0 | 0 | — |
case-08 | pass→pass | 12,553 | 20,201 | +61% | 1 | 1 | 0% | 1,792 | 3,896 | +117% | 0 | 0 | — |
case-09 | pass→fail | 9,173 | 21,197 | +131% | 1 | 1 | 0% | 1,341 | 4,347 | +224% | 0 | 0 | — |
case-10 | fail→fail | 28,626 | 37,773 | +32% | 1 | 1 | 0% | 4,667 | 6,466 | +39% | 0 | 0 | — |
case-11 | pass→pass | 11,512 | 8,165 | -29% | 1 | 1 | 0% | 1,637 | 1,416 | -14% | 0 | 0 | — |
case-12 | pass→pass | 9,145 | 3,410 | -63% | 1 | 1 | 0% | 1,364 | 843 | -38% | 0 | 0 | — |
case-22 | fail→pass | 7,928 | 2,838 | -64% | 1 | 1 | 0% | 1,221 | 734 | -40% | 0 | 0 | — |
case-13 | fail→fail | 3,026 | 24,991 | +726% | 1 | 1 | 0% | 451 | 3,771 | +736% | 0 | 0 | — |
case-14 | pass→pass | 14,147 | 7,982 | -44% | 1 | 1 | 0% | 2,054 | 1,485 | -28% | 0 | 0 | — |
case-15 | pass→pass | 7,523 | 3,921 | -48% | 1 | 1 | 0% | 1,163 | 941 | -19% | 0 | 0 | — |
case-16 | fail→pass | 9,541 | 3,732 | -61% | 1 | 1 | 0% | 1,501 | 882 | -41% | 0 | 0 | — |
case-17 | fail→pass | 13,463 | 9,695 | -28% | 1 | 1 | 0% | 1,819 | 1,721 | -5% | 0 | 0 | — |
case-18 | fail→pass | 16,508 | 3,709 | -78% | 1 | 1 | 0% | 2,332 | 904 | -61% | 0 | 0 | — |
case-19 | fail→pass | 14,728 | 6,120 | -58% | 1 | 1 | 0% | 2,089 | 1,142 | -45% | 0 | 0 | — |
case-20 | fail→fail | 22,750 | 29,939 | +32% | 1 | 1 | 0% | 3,260 | 4,846 | +49% | 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. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.