Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Pathfinder traversal of the knowledge graph starting from a seed entity
.claude/skills/ruvnet-kg-traverse/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -56% | 0% |
Perform pathfinder graph traversal starting from a seed entity. Expands outward through causal edges, scores paths by relevance, and prunes low-similarity branches.
When you need to explore the knowledge graph starting from a specific entity -- finding what depends on it, what it depends on, or discovering indirect relationships. Useful for impact analysis, dependency chains, and understanding code structure.
mcp__plugin_ruflo-core_ruflo__agentdb_hierarchical-recall to look up the target entity by namemcp__plugin_ruflo-core_ruflo__agentdb_causal-edge to find all edges connected to the seed entity, then recursively expand outward to the specified depth (default: 3)cumulative_score = product(edge_weight * keyword_similarity(query, node)) using mcp__plugin_ruflo-core_ruflo__agentdb_pattern-search (the semanticRouter controller is enabled: false in current AgentDB builds; pattern-search is the available substitute and works fine for entity-name + relation-type keyword matches — see ruvnet/ruflo#2049). For higher-fidelity semantic similarity, callers can fall back to mcp__plugin_ruflo-core_ruflo__embeddings_generate + manual cosine, but that's not required for step 3 to function.mcp__plugin_ruflo-core_ruflo__agentdb_context-synthesize to combine the top paths into a coherent summarybashnpx @claude-flow/cli@latest memory search --query "relations for ENTITY_NAME" --namespace knowledge-graph
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 15,533 | 13,925 | -10% | 1 | 1 | 0% | 3,615 | 3,549 | -2% | 0 | 0 | — |
case-21 | pass→pass | 13,987 | 12,805 | -8% | 1 | 1 | 0% | 3,036 | 3,327 | +10% | 0 | 0 | — |
case-01 | fail→fail | 13,856 | 2,729 | -80% | 1 | 1 | 0% | 2,629 | 882 | -66% | 0 | 0 | — |
case-02 | fail→fail | 16,736 | 4,620 | -72% | 1 | 1 | 0% | 3,105 | 876 | -72% | 0 | 0 | — |
case-03 | fail→fail | 16,236 | 5,460 | -66% | 1 | 1 | 0% | 3,305 | 986 | -70% | 0 | 0 | — |
case-04 | fail→pass | 7,467 | 1,358 | -82% | 1 | 1 | 0% | 1,195 | 691 | -42% | 0 | 0 | — |
case-05 | pass→pass | 16,723 | 5,426 | -68% | 1 | 1 | 0% | 3,061 | 1,467 | -52% | 0 | 0 | — |
case-06 | fail→pass | 5,811 | 1,139 | -80% | 1 | 1 | 0% | 920 | 646 | -30% | 0 | 0 | — |
case-07 | pass→pass | 11,076 | 4,947 | -55% | 1 | 1 | 0% | 1,934 | 661 | -66% | 0 | 0 | — |
case-08 | fail→pass | 5,977 | 1,227 | -79% | 1 | 1 | 0% | 997 | 643 | -36% | 0 | 0 | — |
case-09 | fail→pass | 7,646 | 1,746 | -77% | 1 | 1 | 0% | 1,373 | 762 | -45% | 0 | 0 | — |
case-10 | fail→pass | 8,406 | 1,265 | -85% | 1 | 1 | 0% | 1,447 | 638 | -56% | 0 | 0 | — |
case-15 | fail→pass | 5,903 | 1,530 | -74% | 1 | 1 | 0% | 1,033 | 724 | -30% | 0 | 0 | — |
case-11 | fail→pass | 9,380 | 1,518 | -84% | 1 | 1 | 0% | 1,767 | 727 | -59% | 0 | 0 | — |
case-12 | fail→pass | 2,167 | 1,388 | -36% | 1 | 1 | 0% | 295 | 657 | +123% | 0 | 0 | — |
case-13 | fail→pass | 8,416 | 1,395 | -83% | 1 | 1 | 0% | 1,347 | 722 | -46% | 0 | 0 | — |
case-14 | fail→pass | 6,133 | 1,616 | -74% | 1 | 1 | 0% | 1,095 | 739 | -33% | 0 | 0 | — |
case-16 | fail→pass | 7,849 | 1,704 | -78% | 1 | 1 | 0% | 1,307 | 744 | -43% | 0 | 0 | — |
case-17 | fail→pass | 15,023 | 1,252 | -92% | 1 | 1 | 0% | 2,383 | 660 | -72% | 0 | 0 | — |
case-18 | fail→pass | 7,792 | 1,630 | -79% | 1 | 1 | 0% | 1,407 | 707 | -50% | 0 | 0 | — |
case-19 | fail→pass | 11,437 | 3,636 | -68% | 1 | 1 | 0% | 1,839 | 1,097 | -40% | 0 | 0 | — |
case-22 | pass→pass | 11,403 | 10,326 | -9% | 1 | 1 | 0% | 2,921 | 3,161 | +8% | 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 20 counted toward the lift figure. The other 2 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 +64 percentage points is the difference between those two pass rates over the 20 comparable cases. 1 case got worse with the skill loaded, and it is 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.