Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute orchestrate multi-agent systems with handoffs, routing, and workflows across AI providers. Use when building complex AI systems requiring agent collaboration, task delegation, or workflow coordination. Trigger with phrases like "create multi-agent system", "orchestrate agents", or "coordinate agent workflows".
.claude/skills/dicklesworthstone-orchestrating-multi-agent-systems/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | -60% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 30% | 0% |
This skill provides automated assistance for the described functionality.
Before using this skill, ensure you have:
npm install ai)See {baseDir}/references/implementation.md for detailed implementation guide.
See {baseDir}/references/errors.md for comprehensive error handling.
See {baseDir}/references/examples.md for detailed examples.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 19,952 | 19,775 | -1% | 1 | 1 | 0% | 3,011 | 3,223 | +7% | 0 | 0 | — |
case-02 | pass→pass | 11,294 | 10,461 | -7% | 1 | 1 | 0% | 1,289 | 1,680 | +30% | 0 | 0 | — |
case-03 | pass→pass | 17,982 | 16,293 | -9% | 1 | 1 | 0% | 2,538 | 2,602 | +3% | 0 | 0 | — |
case-04 | pass→pass | 39,105 | 7,663 | -80% | 1 | 1 | 0% | 2,602 | 1,653 | -36% | 0 | 0 | — |
case-05 | pass→pass | 17,829 | 11,787 | -34% | 1 | 1 | 0% | 2,464 | 1,529 | -38% | 0 | 0 | — |
case-06 | pass→pass | 23,532 | 19,940 | -15% | 1 | 1 | 0% | 3,515 | 3,062 | -13% | 0 | 0 | — |
case-07 | pass→pass | 13,303 | 17,964 | +35% | 1 | 1 | 0% | 1,393 | 2,790 | +100% | 0 | 0 | — |
case-08 | pass→pass | 17,941 | 17,019 | -5% | 1 | 1 | 0% | 2,077 | 2,437 | +17% | 0 | 0 | — |
case-09 | pass→pass | 20,759 | 20,339 | -2% | 1 | 1 | 0% | 2,543 | 3,084 | +21% | 0 | 0 | — |
case-10 | pass→pass | 37,718 | 19,462 | -48% | 1 | 1 | 0% | 3,141 | 3,406 | +8% | 0 | 0 | — |
case-11 | pass→pass | 23,167 | 18,420 | -20% | 1 | 1 | 0% | 3,370 | 3,050 | -9% | 0 | 0 | — |
case-12 | pass→pass | 24,382 | 20,264 | -17% | 1 | 1 | 0% | 3,387 | 3,145 | -7% | 0 | 0 | — |
case-13 | fail→pass | 15,538 | 7,170 | -54% | 1 | 1 | 0% | 1,631 | 658 | -60% | 0 | 0 | — |
case-14 | fail→pass | 12,111 | 6,442 | -47% | 1 | 1 | 0% | 1,254 | 515 | -59% | 0 | 0 | — |
case-15 | fail→pass | 14,329 | 7,199 | -50% | 1 | 1 | 0% | 1,657 | 717 | -57% | 0 | 0 | — |
case-16 | pass→pass | 14,606 | 8,803 | -40% | 1 | 1 | 0% | 1,747 | 997 | -43% | 0 | 0 | — |
case-17 | pass→pass | 17,451 | 12,356 | -29% | 1 | 1 | 0% | 2,383 | 1,909 | -20% | 0 | 0 | — |
case-18 | pass→pass | 18,312 | 20,214 | +10% | 1 | 1 | 0% | 2,064 | 2,494 | +21% | 0 | 0 | — |
case-19 | pass→pass | 10,843 | 6,977 | -36% | 1 | 1 | 0% | 948 | 702 | -26% | 0 | 0 | — |
case-20 | pass→pass | 18,146 | 16,252 | -10% | 1 | 1 | 0% | 1,976 | 2,257 | +14% | 0 | 0 | — |
case-21 | fail→fail | 22,608 | 9,458 | -58% | 1 | 1 | 0% | 2,956 | 1,147 | -61% | 0 | 0 | — |
case-22 | pass→pass | 14,318 | 12,809 | -11% | 1 | 1 | 0% | 1,559 | 695 | -55% | 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 +14 percentage points is the difference between those two pass rates over the 22 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.