Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Wave Accounting toolkit is not currently available as a native integration. No Wave-specific tools were found in the Composio platform. This skill is a placeholder pending future integration.
.claude/skills/openteams-lab-wave-accounting-automation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -68% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -34% | 0% |
> Note: The Wave Accounting toolkit (wave_accounting) does not currently have native tools available in the Composio platform. Searches for Wave Accounting-specific tools return results from other accounting/invoicing platforms (Stripe, Zoho Invoice) instead.
Toolkit docs: composio.dev/toolkits/wave_accounting
This integration is not yet available with native Wave Accounting tools. When Wave Accounting tools become available in Composio, this skill file will be updated with real tool slugs, workflows, and pitfalls.
For accounting and invoicing automation needs, consider these alternatives that are available today:
https://rube.app/mcp
Powered by Composio
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-19 | fail→fail | 10,201 | 4,811 | -53% | 1 | 1 | 0% | 1,514 | 1,019 | -33% | 0 | 0 | — |
case-01 | fail→pass | 10,374 | 3,137 | -70% | 1 | 1 | 0% | 1,587 | 764 | -52% | 0 | 0 | — |
case-02 | fail→fail | 6,699 | 2,852 | -57% | 1 | 1 | 0% | 974 | 681 | -30% | 0 | 0 | — |
case-03 | fail→pass | 17,097 | 4,633 | -73% | 1 | 1 | 0% | 2,565 | 993 | -61% | 0 | 0 | — |
case-04 | fail→pass | 9,996 | 3,660 | -63% | 1 | 1 | 0% | 1,544 | 887 | -43% | 0 | 0 | — |
case-05 | fail→pass | 8,983 | 1,616 | -82% | 1 | 1 | 0% | 1,531 | 484 | -68% | 0 | 0 | — |
case-06 | fail→pass | 4,620 | 1,964 | -57% | 1 | 1 | 0% | 735 | 484 | -34% | 0 | 0 | — |
case-07 | pass→pass | 9,936 | 4,680 | -53% | 1 | 1 | 0% | 1,471 | 981 | -33% | 0 | 0 | — |
case-08 | pass→pass | 11,866 | 3,976 | -66% | 1 | 1 | 0% | 1,837 | 819 | -55% | 0 | 0 | — |
case-09 | pass→pass | 9,277 | 5,571 | -40% | 1 | 1 | 0% | 1,487 | 1,077 | -28% | 0 | 0 | — |
case-10 | pass→pass | 10,448 | 2,274 | -78% | 1 | 1 | 0% | 1,564 | 580 | -63% | 0 | 0 | — |
case-11 | fail→pass | 7,218 | 3,803 | -47% | 1 | 1 | 0% | 1,158 | 783 | -32% | 0 | 0 | — |
case-12 | fail→pass | 12,963 | 6,299 | -51% | 1 | 1 | 0% | 2,475 | 1,224 | -51% | 0 | 0 | — |
case-13 | fail→pass | 13,048 | 3,020 | -77% | 1 | 1 | 0% | 2,123 | 712 | -66% | 0 | 0 | — |
case-14 | fail→pass | 12,574 | 3,368 | -73% | 1 | 1 | 0% | 2,067 | 741 | -64% | 0 | 0 | — |
case-15 | fail→pass | 13,023 | 5,170 | -60% | 1 | 1 | 0% | 2,213 | 1,011 | -54% | 0 | 0 | — |
case-16 | pass→pass | 9,026 | 3,144 | -65% | 1 | 1 | 0% | 1,350 | 770 | -43% | 0 | 0 | — |
case-17 | fail→fail | 12,564 | 8,107 | -35% | 1 | 1 | 0% | 2,075 | 1,525 | -27% | 0 | 0 | — |
case-18 | fail→pass | 5,903 | 2,848 | -52% | 1 | 1 | 0% | 964 | 672 | -30% | 0 | 0 | — |
case-20 | pass→pass | 8,995 | 7,859 | -13% | 1 | 1 | 0% | 1,521 | 1,654 | +9% | 0 | 0 | — |
case-21 | pass→pass | 7,557 | 6,159 | -18% | 1 | 1 | 0% | 1,236 | 1,206 | -2% | 0 | 0 | — |
case-22 | pass→pass | 11,775 | 7,816 | -34% | 1 | 1 | 0% | 1,921 | 1,516 | -21% | 0 | 0 | — |
case-23 | pass→pass | 9,148 | 3,381 | -63% | 1 | 1 | 0% | 1,421 | 737 | -48% | 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 +48 percentage points is the difference between those two pass rates over the 23 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.