Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Component-level surgical operations (subtract/multiply/divide/unify/redirect) from Systematic Inventive Thinking (SIT).
.claude/skills/yogsoth-ai-component-surgery/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 59% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 155% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 56% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -72% | 0% |
Apply SIT operators at the component level to create structural innovations through surgical modification.
| Resource | Target | Current | % | |----------|--------|---------|---| | web-search | 20 | 0 | 0% | | web-research | 5 | 0 | 0% | | paper-overview | 15 | 0 | 0% | | paper-search | 10 | 0 | 0% | | paper-research | 3 | 0 | 0% |
Cannot exit strategy until ≥80% of each budget line is consumed OR yield targets are met with justification for remaining budget.
| Tactic | Role | |--------|------| | component-decomposition | Decompose system into operable components | | combination-mapping | Map component × operator combinations |
| SOP | Role | |-----|------| | surgery-operation | Execute subtract/multiply/divide/unify/redirect | | function-model-construction | Build functional model before surgery | | recombination-generation | Reassemble post-surgery fragments | | structural-synthesis | Synthesize surgical outcomes |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | component-decomposition | Decompose system into functional components, identify dependencies, and surface trimming candidates. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | recombination-generation | Reassemble decomposed system fragments into novel structural arrangements that create emergent value. | | structural-synthesis | Synthesize all structural transformation outputs into a coherent, ranked idea report with lineage tracking. | | surgery-operation | Execute component surgery operations (subtract/multiply/divide/unify/redirect) from Systematic Inventive Thinking. |
<!-- 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-01 | fail→fail | 39,987 | 7,983 | -80% | 1 | 1 | 0% | 6,221 | 1,078 | -83% | 0 | 0 | — |
case-02 | fail→fail | 35,714 | 7,962 | -78% | 1 | 1 | 0% | 5,840 | 1,013 | -83% | 0 | 0 | — |
case-03 | pass→fail | 22,316 | 34,283 | +54% | 1 | 1 | 0% | 4,313 | 6,719 | +56% | 0 | 0 | — |
case-04 | pass→fail | 23,419 | 8,692 | -63% | 1 | 1 | 0% | 3,841 | 1,085 | -72% | 0 | 0 | — |
case-05 | pass→fail | 17,718 | 27,799 | +57% | 1 | 1 | 0% | 3,022 | 4,253 | +41% | 0 | 0 | — |
case-06 | pass→pass | 18,216 | 28,353 | +56% | 1 | 1 | 0% | 3,434 | 5,328 | +55% | 0 | 0 | — |
case-07 | fail→fail | 19,620 | 31,356 | +60% | 1 | 1 | 0% | 2,573 | 5,415 | +110% | 0 | 0 | — |
case-08 | fail→fail | 15,520 | 31,787 | +105% | 1 | 1 | 0% | 2,094 | 5,093 | +143% | 0 | 0 | — |
case-09 | fail→pass | 21,101 | 25,636 | +21% | 1 | 1 | 0% | 2,447 | 4,369 | +79% | 0 | 0 | — |
case-10 | pass→pass | 17,103 | 15,632 | -9% | 1 | 1 | 0% | 2,413 | 2,638 | +9% | 0 | 0 | — |
case-11 | pass→fail | 11,646 | 10,146 | -13% | 1 | 1 | 0% | 1,584 | 1,153 | -27% | 0 | 0 | — |
case-12 | pass→fail | 17,721 | 9,893 | -44% | 1 | 1 | 0% | 2,515 | 1,034 | -59% | 0 | 0 | — |
case-13 | pass→fail | 19,546 | 40,017 | +105% | 1 | 1 | 0% | 2,819 | 5,786 | +105% | 0 | 0 | — |
case-14 | fail→pass | 17,329 | 26,655 | +54% | 1 | 1 | 0% | 2,626 | 4,185 | +59% | 0 | 0 | — |
case-15 | fail→fail | 13,795 | 9,723 | -30% | 1 | 1 | 0% | 2,225 | 2,015 | -9% | 0 | 0 | — |
case-16 | pass→pass | 19,038 | 24,396 | +28% | 1 | 1 | 0% | 2,620 | 4,280 | +63% | 0 | 0 | — |
case-17 | pass→pass | 18,438 | 39,617 | +115% | 1 | 1 | 0% | 2,514 | 6,711 | +167% | 0 | 0 | — |
case-18 | fail→fail | 16,230 | 38,398 | +137% | 1 | 1 | 0% | 2,325 | 6,713 | +189% | 0 | 0 | — |
case-19 | pass→pass | 21,126 | 33,758 | +60% | 1 | 1 | 0% | 3,052 | 5,581 | +83% | 0 | 0 | — |
case-20 | fail→pass | 11,380 | 22,031 | +94% | 1 | 1 | 0% | 1,548 | 3,949 | +155% | 0 | 0 | — |
case-21 | fail→fail | 40,934 | 41,179 | +1% | 1 | 1 | 0% | 6,167 | 6,205 | +1% | 0 | 0 | — |
case-22 | pass→pass | 14,198 | 8,629 | -39% | 1 | 1 | 0% | 2,006 | 1,824 | -9% | 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 16 counted toward the lift figure. The other 6 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 -14 percentage points is the difference between those two pass rates over the 16 comparable cases. 6 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.