Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Manage RVF (Ruflo Vector Format) files for portable agent memory and cross-platform transfer
.claude/skills/ruvnet-rvf-manage/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -67% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -48% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -68% | 0% |
Manage RVF files for portable, transferable agent memory.
When you need to export agent memory to RVF format for backup, transfer between projects, or share knowledge between teams.
mcp__plugin_ruflo-core_ruflo__memory_list to see all stored memoriesmcp__plugin_ruflo-core_ruflo__hooks_transfer tool with store action to export patternsmcp__plugin_ruflo-core_ruflo__memory_import_claude to import from Claude Code memoriesmcp__plugin_ruflo-core_ruflo__memory_migrate for format upgradesmcp__plugin_ruflo-core_ruflo__memory_stats for storage metricsRVF (Ruflo Vector Format) stores:
bashnpx @claude-flow/cli@latest hooks transfer store --pattern "project-knowledge" npx @claude-flow/cli@latest hooks transfer from-project --source /path/to/other/project
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | fail→pass | 6,231 | 1,653 | -73% | 1 | 1 | 0% | 1,139 | 634 | -44% | 0 | 0 | — |
case-01 | fail→fail | 8,812 | 5,279 | -40% | 1 | 1 | 0% | 1,225 | 367 | -70% | 0 | 0 | — |
case-02 | fail→fail | 11,599 | 2,504 | -78% | 1 | 1 | 0% | 2,121 | 757 | -64% | 0 | 0 | — |
case-03 | fail→fail | 9,416 | 4,076 | -57% | 1 | 1 | 0% | 1,591 | 698 | -56% | 0 | 0 | — |
case-04 | fail→pass | 9,200 | 4,618 | -50% | 1 | 1 | 0% | 1,723 | 577 | -67% | 0 | 0 | — |
case-05 | fail→pass | 5,742 | 1,704 | -70% | 1 | 1 | 0% | 981 | 667 | -32% | 0 | 0 | — |
case-07 | fail→pass | 9,690 | 3,605 | -63% | 1 | 1 | 0% | 2,062 | 1,069 | -48% | 0 | 0 | — |
case-08 | fail→pass | 10,784 | 5,765 | -47% | 1 | 1 | 0% | 2,047 | 657 | -68% | 0 | 0 | — |
case-09 | pass→pass | 9,209 | 1,487 | -84% | 1 | 1 | 0% | 1,583 | 588 | -63% | 0 | 0 | — |
case-10 | pass→pass | 11,776 | 4,645 | -61% | 1 | 1 | 0% | 2,199 | 1,215 | -45% | 0 | 0 | — |
case-11 | pass→pass | 15,427 | 5,889 | -62% | 1 | 1 | 0% | 2,524 | 1,351 | -46% | 0 | 0 | — |
case-12 | pass→pass | 16,234 | 8,397 | -48% | 1 | 1 | 0% | 2,801 | 1,850 | -34% | 0 | 0 | — |
case-13 | fail→pass | 7,392 | 3,717 | -50% | 1 | 1 | 0% | 1,432 | 1,053 | -26% | 0 | 0 | — |
case-14 | fail→pass | 7,665 | 1,861 | -76% | 1 | 1 | 0% | 1,772 | 695 | -61% | 0 | 0 | — |
case-15 | fail→pass | 5,412 | 1,761 | -67% | 1 | 1 | 0% | 945 | 635 | -33% | 0 | 0 | — |
case-16 | fail→pass | 8,571 | 4,209 | -51% | 1 | 1 | 0% | 1,557 | 786 | -50% | 0 | 0 | — |
case-17 | fail→pass | 6,095 | 5,046 | -17% | 1 | 1 | 0% | 1,092 | 565 | -48% | 0 | 0 | — |
case-18 | fail→pass | 14,084 | 1,784 | -87% | 1 | 1 | 0% | 2,486 | 613 | -75% | 0 | 0 | — |
case-19 | fail→pass | 4,091 | 1,664 | -59% | 1 | 1 | 0% | 766 | 607 | -21% | 0 | 0 | — |
case-20 | pass→pass | 3,554 | 3,935 | +11% | 1 | 1 | 0% | 590 | 1,014 | +72% | 0 | 0 | — |
case-21 | pass→pass | 3,541 | 3,971 | +12% | 1 | 1 | 0% | 650 | 838 | +29% | 0 | 0 | — |
case-22 | pass→pass | 3,447 | 2,827 | -18% | 1 | 1 | 0% | 548 | 789 | +44% | 0 | 0 | — |
case-23 | fail→pass | 4,877 | 4,337 | -11% | 1 | 1 | 0% | 1,066 | 649 | -39% | 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, and 22 counted toward the lift figure. The other 1 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 +57 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.
Other measured skills in the registry, with their headline benchmark lift.