Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Orchestrate the excursion sequence — departure into unrelated domain, force-fit discoveries back to problem, launch springboard ideas.
.claude/skills/yogsoth-ai-excursion-orchestration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 63% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -62% | 0% |
Orchestrate the full excursion sequence from departure through force-fit to springboard launch.
Leave the problem entirely. Select an unrelated domain (nature, art, sport, cooking, music, etc.) and explore it deeply using excursion-departure SOP. The key is genuine immersion — not looking for connections yet.
Take the most interesting discoveries from the excursion domain and deliberately, forcefully connect them back to the original problem using force-fit SOP. Accept awkward connections — they often yield the best insights.
Convert the force-fitted connections into "I wish..." or "How to..." springboard statements using springboard-launch SOP. Develop the most promising springboards into concrete solution concepts.
| Metric | Floor | |--------|-------| | Excursion discoveries | ≥5 | | Force-fitted ideas | ≥3 | | Springboard statements | ≥3 | | Concrete solutions | ≥2 |
| SOP | Role | |-----|------| | excursion-departure | Stage 1 — domain departure and exploration | | force-fit | Stage 2 — force connections back to problem | | springboard-launch | Stage 3 — convert to concrete solutions | | direct-analogy-generation | Support — find analogies in excursion domain | | synectics-synthesis | Post — synthesize excursion outputs |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | excursion-departure | Leave the problem entirely and explore an unrelated domain. Produces excursion domain discoveries for later force-fitting. | | force-fit | Force-fit excursion discoveries back to the original problem. Deliberately create connections between unrelated findings and the challenge. | | springboard-launch | Convert analogy insights into concrete feasible solutions. Transform abstract connections into actionable mechanisms. |
<!-- 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-05 | fail→fail | 14,157 | 22,697 | +60% | 1 | 1 | 0% | 2,089 | 3,784 | +81% | 0 | 0 | — |
case-06 | pass→pass | 15,774 | 19,350 | +23% | 1 | 1 | 0% | 2,341 | 3,334 | +42% | 0 | 0 | — |
case-23 | pass→pass | 17,218 | 18,569 | +8% | 1 | 1 | 0% | 2,715 | 3,398 | +25% | 0 | 0 | — |
case-20 | pass→fail | 11,738 | 28,072 | +139% | 1 | 1 | 0% | 1,870 | 4,705 | +152% | 0 | 0 | — |
case-21 | pass→fail | 18,921 | 29,004 | +53% | 1 | 1 | 0% | 2,781 | 5,034 | +81% | 0 | 0 | — |
case-01 | fail→fail | 24,290 | 49,844 | +105% | 1 | 1 | 0% | 3,701 | 4,987 | +35% | 0 | 0 | — |
case-02 | fail→fail | 26,815 | 21,098 | -21% | 1 | 1 | 0% | 4,165 | 3,591 | -14% | 0 | 0 | — |
case-03 | fail→pass | 18,481 | 28,195 | +53% | 1 | 1 | 0% | 2,994 | 4,875 | +63% | 0 | 0 | — |
case-04 | fail→fail | 11,844 | 14,977 | +26% | 1 | 1 | 0% | 1,721 | 2,746 | +60% | 0 | 0 | — |
case-07 | fail→fail | 13,690 | 19,073 | +39% | 1 | 1 | 0% | 1,998 | 3,365 | +68% | 0 | 0 | — |
case-08 | fail→pass | 15,226 | 23,476 | +54% | 1 | 1 | 0% | 2,421 | 3,976 | +64% | 0 | 0 | — |
case-09 | fail→fail | 19,683 | 23,955 | +22% | 1 | 1 | 0% | 2,949 | 3,923 | +33% | 0 | 0 | — |
case-10 | fail→fail | 7,861 | 12,670 | +61% | 1 | 1 | 0% | 1,181 | 2,243 | +90% | 0 | 0 | — |
case-11 | pass→pass | 9,111 | 1,585 | -83% | 1 | 1 | 0% | 1,298 | 698 | -46% | 0 | 0 | — |
case-12 | pass→pass | 6,479 | 1,849 | -71% | 1 | 1 | 0% | 1,028 | 742 | -28% | 0 | 0 | — |
case-13 | fail→pass | 8,536 | 2,050 | -76% | 1 | 1 | 0% | 1,357 | 786 | -42% | 0 | 0 | — |
case-14 | pass→pass | 8,726 | 1,867 | -79% | 1 | 1 | 0% | 1,457 | 756 | -48% | 0 | 0 | — |
case-22 | pass→fail | 15,205 | 29,796 | +96% | 1 | 1 | 0% | 2,710 | 5,034 | +86% | 0 | 0 | — |
case-15 | fail→pass | 9,971 | 1,737 | -83% | 1 | 1 | 0% | 1,598 | 773 | -52% | 0 | 0 | — |
case-16 | fail→pass | 11,423 | 2,053 | -82% | 1 | 1 | 0% | 1,894 | 728 | -62% | 0 | 0 | — |
case-17 | pass→pass | 13,433 | 7,590 | -43% | 1 | 1 | 0% | 2,009 | 1,606 | -20% | 0 | 0 | — |
case-18 | fail→fail | 12,021 | 6,685 | -44% | 1 | 1 | 0% | 1,874 | 1,603 | -14% | 0 | 0 | — |
case-19 | pass→pass | 16,802 | 21,113 | +26% | 1 | 1 | 0% | 2,596 | 3,690 | +42% | 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. 23 cases were attempted. The headline lift of +9 percentage points is the difference between those two pass rates over the 23 comparable cases. 3 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.