Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when the user explicitly asks to diagnose or improve Codex skill/plugin routing, determine which skill or plugin should handle a request, refresh or check a skill registry, investigate why a skill/plugin did not trigger, or review skill descriptions for trigger quality. Do not use for ordinary task execution when a concrete domain skill clearly applies, unless routing is ambiguous or the user asks for routing.
.claude/skills/hashgraph-online-skill-router/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -14% | 0% |
Use this skill as a diagnostic and routing-assistance layer. It does not replace Codex native skill discovery, Superpowers process skills, or subagent orchestration. It helps decide which capabilities are relevant, why they are relevant, and when not to use them.
Use this skill when the user asks things like:
Do not use this skill for simple tasks with an obvious primary skill:
If the task is complex or ambiguous, perform a brief routing check and then continue with the selected primary skill.
textRecommended skill: - ... Helper skills: - ... Why: - ... Do not use: - ... Needs confirmation: - ...
Use scripts/build_registry.py only when the user explicitly asks to refresh the registry or when the registry is stale, missing, or inconsistent.
Use scripts/route_request.py --check-registry to inspect registry health.
See references/capability-card-schema.md for card fields and provenance requirements.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 18,857 | 15,087 | -20% | 1 | 1 | 0% | 1,855 | 2,127 | +15% | 0 | 0 | — |
case-02 | fail→pass | 16,858 | 13,768 | -18% | 1 | 1 | 0% | 2,618 | 2,138 | -18% | 0 | 0 | — |
case-03 | fail→pass | 13,189 | 13,614 | +3% | 1 | 1 | 0% | 2,165 | 1,989 | -8% | 0 | 0 | — |
case-04 | pass→fail | 14,601 | 13,453 | -8% | 1 | 1 | 0% | 1,341 | 2,089 | +56% | 0 | 0 | — |
case-05 | fail→fail | 56,282 | 27,453 | -51% | 1 | 1 | 0% | 8,210 | 5,203 | -37% | 0 | 0 | — |
case-06 | fail→fail | 7,856 | 8,803 | +12% | 1 | 1 | 0% | 313 | 1,038 | +232% | 0 | 0 | — |
case-07 | pass→fail | 43,806 | 25,923 | -41% | 1 | 1 | 0% | 6,503 | 2,442 | -62% | 0 | 0 | — |
case-08 | fail→pass | 14,957 | 12,860 | -14% | 1 | 1 | 0% | 2,137 | 1,714 | -20% | 0 | 0 | — |
case-09 | fail→pass | 23,717 | 13,330 | -44% | 1 | 1 | 0% | 2,838 | 2,430 | -14% | 0 | 0 | — |
case-10 | fail→pass | 19,497 | 6,837 | -65% | 1 | 1 | 0% | 2,114 | 1,730 | -18% | 0 | 0 | — |
case-16 | fail→pass | 24,730 | 11,164 | -55% | 1 | 1 | 0% | 2,688 | 2,051 | -24% | 0 | 0 | — |
case-11 | fail→pass | 16,442 | 1,692 | -90% | 1 | 1 | 0% | 2,492 | 892 | -64% | 0 | 0 | — |
case-12 | pass→pass | 18,096 | 2,552 | -86% | 1 | 1 | 0% | 2,275 | 924 | -59% | 0 | 0 | — |
case-13 | fail→pass | 10,707 | 6,237 | -42% | 1 | 1 | 0% | 1,791 | 789 | -56% | 0 | 0 | — |
case-14 | fail→pass | 31,447 | 18,026 | -43% | 1 | 1 | 0% | 3,704 | 2,920 | -21% | 0 | 0 | — |
case-15 | fail→pass | 17,400 | 22,620 | +30% | 1 | 1 | 0% | 2,366 | 2,996 | +27% | 0 | 0 | — |
case-17 | fail→pass | 16,985 | 21,903 | +29% | 1 | 1 | 0% | 2,988 | 3,549 | +19% | 0 | 0 | — |
case-18 | fail→pass | 8,397 | 12,395 | +48% | 1 | 1 | 0% | 1,308 | 1,640 | +25% | 0 | 0 | — |
case-19 | pass→pass | 10,463 | 9,406 | -10% | 1 | 1 | 0% | 1,605 | 1,278 | -20% | 0 | 0 | — |
case-20 | fail→pass | 14,298 | 7,761 | -46% | 1 | 1 | 0% | 1,781 | 1,086 | -39% | 0 | 0 | — |
case-21 | fail→pass | 21,114 | 16,615 | -21% | 1 | 1 | 0% | 2,567 | 2,471 | -4% | 0 | 0 | — |
case-22 | fail→pass | 13,458 | 16,827 | +25% | 1 | 1 | 0% | 2,281 | 2,315 | +1% | 0 | 0 | — |
case-23 | pass→pass | 10,860 | 12,909 | +19% | 1 | 1 | 0% | 1,657 | 1,913 | +15% | 0 | 0 | — |
case-24 | fail→pass | 18,145 | 1,778 | -90% | 1 | 1 | 0% | 2,458 | 852 | -65% | 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 +63 percentage points is the difference between those two pass rates over the 24 comparable cases. 3 cases got worse with the skill loaded, and they are 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.