Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Create a new weft workflow template interactively. Use only when user types /wf-new-template.
.claude/skills/ccplugins-wf-new-template/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -40% | 0% |
Build a custom workflow template through conversation.
$ARGUMENTS
Walk the user through creating a template step by step:
If no name in arguments, ask: "What should this workflow be called? (lowercase, hyphens ok)"
Ask: "One-line description of what this workflow does?"
Ask: "List the steps in order. You can describe them naturally and I'll structure them."
Example user input: "first plan the work, then implement it, run tests, do a code review, then push" → Parse into: plan, implement, test, review, push
For each step, ask if the default on_fail: block is ok, or if they want:
retry — auto-retry once on failurecontinue — skip failed step and move onblock — halt until manual retry (default)Suggest sensible defaults: test steps → retry, optional steps → continue.
Ask: "Should any steps block certain commands? For example, blocking git push until tests pass."
If yes, collect:
git push, git commit)Build the template JSON and show a preview using:
bashecho '<json>' | python3 "${CLAUDE_PLUGIN_ROOT}/core/cli.py" save-template
Before saving, show the full template and ask: "Look good? Save it?"
bashecho '<template_json>' | python3 "${CLAUDE_PLUGIN_ROOT}/core/cli.py" save-template
The template is saved to .claude/weft/templates/<name>.json in the project directory.
Tell the user: "Template saved! Start it with /wf-start <name>"
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 5,398 | 2,603 | -52% | 1 | 1 | 0% | 749 | 777 | +4% | 0 | 0 | — |
case-02 | fail→fail | 30,924 | 3,101 | -90% | 1 | 1 | 0% | 3,766 | 870 | -77% | 0 | 0 | — |
case-03 | fail→pass | 4,993 | 3,274 | -34% | 1 | 1 | 0% | 836 | 850 | +2% | 0 | 0 | — |
case-04 | fail→pass | 6,632 | 3,127 | -53% | 1 | 1 | 0% | 1,020 | 861 | -16% | 0 | 0 | — |
case-05 | pass→pass | 14,448 | 6,172 | -57% | 1 | 1 | 0% | 1,833 | 1,337 | -27% | 0 | 0 | — |
case-06 | pass→pass | 13,672 | 12,848 | -6% | 1 | 1 | 0% | 2,030 | 1,846 | -9% | 0 | 0 | — |
case-07 | fail→pass | 13,737 | 6,482 | -53% | 1 | 1 | 0% | 1,806 | 1,313 | -27% | 0 | 0 | — |
case-08 | pass→pass | 12,764 | 6,808 | -47% | 1 | 1 | 0% | 1,727 | 1,281 | -26% | 0 | 0 | — |
case-09 | pass→pass | 10,833 | 5,395 | -50% | 1 | 1 | 0% | 1,692 | 1,222 | -28% | 0 | 0 | — |
case-10 | fail→pass | 11,873 | 3,180 | -73% | 1 | 1 | 0% | 1,667 | 861 | -48% | 0 | 0 | — |
case-11 | fail→pass | 8,428 | 1,973 | -77% | 1 | 1 | 0% | 1,263 | 754 | -40% | 0 | 0 | — |
case-12 | fail→pass | 8,858 | 2,558 | -71% | 1 | 1 | 0% | 1,440 | 796 | -45% | 0 | 0 | — |
case-13 | pass→pass | 21,477 | 6,689 | -69% | 1 | 1 | 0% | 3,148 | 1,467 | -53% | 0 | 0 | — |
case-14 | fail→pass | 11,667 | 4,806 | -59% | 1 | 1 | 0% | 1,653 | 1,067 | -35% | 0 | 0 | — |
case-15 | fail→pass | 12,738 | 4,752 | -63% | 1 | 1 | 0% | 1,849 | 1,079 | -42% | 0 | 0 | — |
case-16 | pass→pass | 11,864 | 4,079 | -66% | 1 | 1 | 0% | 1,634 | 809 | -50% | 0 | 0 | — |
case-17 | fail→pass | 8,945 | 2,831 | -68% | 1 | 1 | 0% | 1,146 | 830 | -28% | 0 | 0 | — |
case-18 | fail→fail | 12,671 | 2,763 | -78% | 1 | 1 | 0% | 1,577 | 786 | -50% | 0 | 0 | — |
case-19 | fail→pass | 12,452 | 3,398 | -73% | 1 | 1 | 0% | 1,776 | 895 | -50% | 0 | 0 | — |
case-20 | fail→pass | 5,761 | 2,888 | -50% | 1 | 1 | 0% | 763 | 778 | +2% | 0 | 0 | — |
case-21 | fail→pass | 13,910 | 3,472 | -75% | 1 | 1 | 0% | 2,146 | 976 | -55% | 0 | 0 | — |
case-22 | pass→pass | 17,175 | 2,219 | -87% | 1 | 1 | 0% | 2,463 | 765 | -69% | 0 | 0 | — |
case-23 | pass→pass | 12,279 | 5,312 | -57% | 1 | 1 | 0% | 1,940 | 1,269 | -35% | 0 | 0 | — |
case-24 | pass→fail | 13,339 | 3,137 | -76% | 1 | 1 | 0% | 1,737 | 876 | -50% | 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. 24 cases were attempted. The headline lift of +46 percentage points is the difference between those two pass rates over the 24 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.