Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Search and install AI assets from TokRepo when a user asks to find, discover, or install Codex skills, MCP servers, prompts, cursor rules, or workflows.
.claude/skills/hashgraph-online-tokrepo-search/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -57% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -70% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -74% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -66% | 0% |
Use this skill when the user needs to discover installable AI assets such as Codex skills, MCP servers, prompts, cursor rules, or workflows.
bashnpx tokrepo search "<query>"
Examples:
bashnpx tokrepo search "mcp database" npx tokrepo search "codex skill github" npx tokrepo search "cursor rules react"
bashnpx tokrepo install <uuid-or-name>
TokRepo can surface:
bashnpx tokrepo search ""
npx tokrepo install over recreating files by hand| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 13,407 | 15,133 | +13% | 1 | 1 | 0% | 1,508 | 422 | -72% | 0 | 0 | — |
case-02 | fail→fail | 11,108 | 10,557 | -5% | 1 | 1 | 0% | 1,617 | 421 | -74% | 0 | 0 | — |
case-03 | fail→fail | 18,344 | 13,121 | -28% | 1 | 1 | 0% | 2,602 | 516 | -80% | 0 | 0 | — |
case-04 | pass→pass | 13,340 | 4,098 | -69% | 1 | 1 | 0% | 1,180 | 927 | -21% | 0 | 0 | — |
case-05 | pass→pass | 16,077 | 14,232 | -11% | 1 | 1 | 0% | 1,859 | 1,483 | -20% | 0 | 0 | — |
case-06 | pass→pass | 7,038 | 7,785 | +11% | 1 | 1 | 0% | 273 | 539 | +97% | 0 | 0 | — |
case-07 | fail→fail | 18,456 | 15,062 | -18% | 1 | 1 | 0% | 2,254 | 470 | -79% | 0 | 0 | — |
case-08 | fail→fail | 10,296 | 9,058 | -12% | 1 | 1 | 0% | 1,451 | 603 | -58% | 0 | 0 | — |
case-09 | fail→pass | 10,791 | 7,900 | -27% | 1 | 1 | 0% | 1,420 | 621 | -56% | 0 | 0 | — |
case-10 | fail→pass | 7,861 | 1,477 | -81% | 1 | 1 | 0% | 990 | 426 | -57% | 0 | 0 | — |
case-11 | fail→pass | 33,304 | 2,540 | -92% | 1 | 1 | 0% | 2,097 | 624 | -70% | 0 | 0 | — |
case-12 | fail→pass | 14,969 | 2,554 | -83% | 1 | 1 | 0% | 2,400 | 625 | -74% | 0 | 0 | — |
case-13 | fail→pass | 17,927 | 8,206 | -54% | 1 | 1 | 0% | 2,126 | 724 | -66% | 0 | 0 | — |
case-14 | fail→pass | 20,122 | 10,043 | -50% | 1 | 1 | 0% | 2,105 | 1,014 | -52% | 0 | 0 | — |
case-15 | fail→pass | 28,678 | 1,553 | -95% | 1 | 1 | 0% | 3,161 | 430 | -86% | 0 | 0 | — |
case-16 | fail→pass | 15,008 | 8,001 | -47% | 1 | 1 | 0% | 1,721 | 582 | -66% | 0 | 0 | — |
case-17 | fail→pass | 10,935 | 2,680 | -75% | 1 | 1 | 0% | 1,684 | 566 | -66% | 0 | 0 | — |
case-18 | fail→pass | 16,192 | 2,102 | -87% | 1 | 1 | 0% | 1,494 | 539 | -64% | 0 | 0 | — |
case-19 | fail→pass | 10,964 | 2,540 | -77% | 1 | 1 | 0% | 1,030 | 558 | -46% | 0 | 0 | — |
case-20 | pass→pass | 22,560 | 7,183 | -68% | 1 | 1 | 0% | 1,806 | 590 | -67% | 0 | 0 | — |
case-21 | fail→pass | 12,374 | 2,419 | -80% | 1 | 1 | 0% | 1,197 | 488 | -59% | 0 | 0 | — |
case-22 | pass→pass | 14,827 | 9,804 | -34% | 1 | 1 | 0% | 2,093 | 1,075 | -49% | 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, and 18 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +55 percentage points is the difference between those two pass rates over the 18 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.