Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Configure Claude Octopus — redirects to /octo:setup interactive wizard
.claude/skills/nyldn-sys-configure/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -66% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -50% | 0% |
This skill is an alias for /octo:setup. When triggered, invoke the setup command directly.
Setup uses a separate, private resume receipt. It rechecks readiness before trusting a recorded stage, leaves interrupted human login incomplete, and marks the receipt complete only after strict configuration persistence and readback. The user runs browser-opening authentication in their own terminal. Remote sessions never open it automatically.
Action: Run /octo:setup — the interactive setup wizard handles all configuration:
Do NOT duplicate setup logic here. Read ${HOME}/.claude-octopus/plugin/commands/setup.md and follow it only because the user explicitly invoked this configuration skill.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 13,568 | 17,165 | +27% | 1 | 1 | 0% | 1,310 | 613 | -53% | 0 | 0 | — |
case-02 | fail→pass | 14,624 | 9,043 | -38% | 1 | 1 | 0% | 1,374 | 789 | -43% | 0 | 0 | — |
case-03 | fail→pass | 17,838 | 12,034 | -33% | 1 | 1 | 0% | 1,860 | 1,264 | -32% | 0 | 0 | — |
case-04 | pass→pass | 20,207 | 9,341 | -54% | 1 | 1 | 0% | 2,424 | 766 | -68% | 0 | 0 | — |
case-05 | pass→pass | 20,176 | 9,153 | -55% | 1 | 1 | 0% | 2,077 | 774 | -63% | 0 | 0 | — |
case-06 | fail→pass | 13,403 | 6,897 | -49% | 1 | 1 | 0% | 1,202 | 413 | -66% | 0 | 0 | — |
case-07 | pass→pass | 12,537 | 7,630 | -39% | 1 | 1 | 0% | 1,227 | 408 | -67% | 0 | 0 | — |
case-08 | fail→pass | 8,933 | 6,572 | -26% | 1 | 1 | 0% | 597 | 408 | -32% | 0 | 0 | — |
case-09 | fail→pass | 15,826 | 10,539 | -33% | 1 | 1 | 0% | 1,838 | 922 | -50% | 0 | 0 | — |
case-10 | pass→pass | 16,657 | 7,392 | -56% | 1 | 1 | 0% | 1,745 | 539 | -69% | 0 | 0 | — |
case-11 | fail→pass | 15,263 | 7,424 | -51% | 1 | 1 | 0% | 1,701 | 521 | -69% | 0 | 0 | — |
case-12 | fail→fail | 15,521 | 7,098 | -54% | 1 | 1 | 0% | 1,461 | 497 | -66% | 0 | 0 | — |
case-13 | fail→pass | 17,181 | 6,905 | -60% | 1 | 1 | 0% | 1,767 | 404 | -77% | 0 | 0 | — |
case-14 | fail→pass | 16,117 | 7,364 | -54% | 1 | 1 | 0% | 1,782 | 532 | -70% | 0 | 0 | — |
case-15 | fail→pass | 15,541 | 6,714 | -57% | 1 | 1 | 0% | 1,630 | 442 | -73% | 0 | 0 | — |
case-16 | pass→pass | 15,124 | 7,142 | -53% | 1 | 1 | 0% | 1,430 | 391 | -73% | 0 | 0 | — |
case-17 | fail→fail | 18,520 | 6,912 | -63% | 1 | 1 | 0% | 2,166 | 485 | -78% | 0 | 0 | — |
case-18 | fail→fail | 15,000 | 6,797 | -55% | 1 | 1 | 0% | 1,513 | 389 | -74% | 0 | 0 | — |
case-19 | pass→pass | 19,425 | 7,394 | -62% | 1 | 1 | 0% | 2,193 | 483 | -78% | 0 | 0 | — |
case-20 | pass→pass | 14,058 | 10,429 | -26% | 1 | 1 | 0% | 1,771 | 1,174 | -34% | 0 | 0 | — |
case-21 | pass→pass | 10,311 | 11,934 | +16% | 1 | 1 | 0% | 882 | 1,374 | +56% | 0 | 0 | — |
case-22 | pass→pass | 15,791 | 11,253 | -29% | 1 | 1 | 0% | 1,857 | 1,091 | -41% | 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 +41 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/18/2026 | +48% |
| gemini-3.6-flash | verified | 8/11/2026 | +86% |
Other measured skills in the registry, with their headline benchmark lift.