Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create and execute automated workflows for complex multi-step processes.
.claude/skills/workflow-automation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 143% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -63% | 0% |
Create and execute automated workflows for complex multi-step processes.
bashnpx claude-flow workflow create --name "deploy-flow" --template ci
bashnpx claude-flow workflow execute --name "deploy-flow" --env production
bashnpx claude-flow workflow list
bashnpx claude-flow workflow export --name "deploy-flow" --format yaml
bashnpx claude-flow workflow status --name "deploy-flow"
| Template | Description | |----------|-------------| | ci | Continuous integration pipeline | | deploy | Deployment workflow | | test | Testing workflow | | release | Release automation | | review | Code review workflow |
yamlname: example-workflow steps: - name: analyze agent: researcher task: "Analyze requirements" - name: implement agent: coder depends: [analyze] task: "Implement solution" - name: test agent: tester depends: [implement] task: "Write and run tests"
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 7,550 | 1,860 | -75% | 1 | 1 | 0% | 1,553 | 565 | -64% | 0 | 0 | — |
case-02 | fail→pass | 5,130 | 5,049 | -2% | 1 | 1 | 0% | 492 | 1,197 | +143% | 0 | 0 | — |
case-03 | fail→pass | 6,732 | 1,692 | -75% | 1 | 1 | 0% | 1,267 | 720 | -43% | 0 | 0 | — |
case-04 | fail→pass | 9,486 | 1,839 | -81% | 1 | 1 | 0% | 1,605 | 694 | -57% | 0 | 0 | — |
case-05 | fail→pass | 9,043 | 1,445 | -84% | 1 | 1 | 0% | 1,474 | 548 | -63% | 0 | 0 | — |
case-06 | fail→pass | 8,016 | 2,025 | -75% | 1 | 1 | 0% | 1,381 | 653 | -53% | 0 | 0 | — |
case-07 | fail→pass | 4,108 | 3,141 | -24% | 1 | 1 | 0% | 665 | 924 | +39% | 0 | 0 | — |
case-08 | fail→pass | 6,747 | 1,703 | -75% | 1 | 1 | 0% | 1,343 | 575 | -57% | 0 | 0 | — |
case-09 | fail→pass | 6,713 | 1,354 | -80% | 1 | 1 | 0% | 1,003 | 538 | -46% | 0 | 0 | — |
case-10 | fail→pass | 6,801 | 1,393 | -80% | 1 | 1 | 0% | 1,377 | 542 | -61% | 0 | 0 | — |
case-11 | fail→pass | 9,051 | 3,187 | -65% | 1 | 1 | 0% | 1,714 | 686 | -60% | 0 | 0 | — |
case-12 | fail→pass | 8,042 | 5,350 | -33% | 1 | 1 | 0% | 1,528 | 1,079 | -29% | 0 | 0 | — |
case-13 | fail→fail | 3,980 | 3,046 | -23% | 1 | 1 | 0% | 787 | 955 | +21% | 0 | 0 | — |
case-14 | fail→pass | 4,766 | 2,931 | -39% | 1 | 1 | 0% | 905 | 702 | -22% | 0 | 0 | — |
case-15 | fail→pass | 4,240 | 2,111 | -50% | 1 | 1 | 0% | 633 | 535 | -15% | 0 | 0 | — |
case-16 | fail→pass | 8,483 | 1,944 | -77% | 1 | 1 | 0% | 1,420 | 666 | -53% | 0 | 0 | — |
case-17 | fail→fail | 6,781 | 1,475 | -78% | 1 | 1 | 0% | 1,151 | 536 | -53% | 0 | 0 | — |
case-18 | fail→fail | 13,572 | 2,355 | -83% | 1 | 1 | 0% | 2,403 | 692 | -71% | 0 | 0 | — |
case-19 | fail→fail | 5,170 | 3,566 | -31% | 1 | 1 | 0% | 1,012 | 946 | -7% | 0 | 0 | — |
case-20 | fail→fail | 1,898 | 2,168 | +14% | 1 | 1 | 0% | 304 | 738 | +143% | 0 | 0 | — |
case-21 | fail→fail | 1,960 | 2,037 | +4% | 1 | 1 | 0% | 314 | 706 | +125% | 0 | 0 | — |
case-22 | fail→fail | 2,077 | 3,266 | +57% | 1 | 1 | 0% | 290 | 844 | +191% | 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 +68 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.