Kg Traverse
- 633 installs
- 67k repo stars
- Updated August 4, 2026
- ruvnet/ruflo
kg-traverse is a Claude Code skill that performs pathfinder knowledge-graph traversal from a seed entity for developers who need precise retrieval across codebase and documentation context.
About
kg-traverse is an agent retrieval skill from ruvnet/ruflo that performs pathfinder graph traversal starting from a seed entity with an optional --depth limit. It expands outward through causal edges, scores paths by relevance, and prunes low-similarity branches using AgentDB MCP tools including hierarchical-recall, causal-edge, pattern-search, and context-synthesize. Developers reach for kg-traverse when they need to explore what depends on an entity, what it depends on, or discover indirect relationships across codebases, documentation, and project context. The skill reduces hallucination risk by grounding agent answers in graph-linked facts rather than isolated file snippets.
- Traverses knowledge graphs built from repositories and docs
- Enables structured retrieval for agentic workflows
- Reduces hallucinations by grounding responses in graph relationships
- Works with any MCP-compatible knowledge graph server
- Supports complex multi-hop queries across entities and relations
Kg Traverse by the numbers
- 633 all-time installs (skills.sh)
- +10 installs in the week ending Jul 26, 2026 (Skillselion tracking)
- Ranked #1,525 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/ruvnet/ruflo --skill kg-traverseAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 633 |
|---|---|
| repo stars | ★ 67k |
| Last updated | August 4, 2026 |
| Repository | ruvnet/ruflo ↗ |
How do agents query knowledge graphs from a seed entity?
Query and navigate knowledge graphs created from codebases, documentation, and project context so agents can retrieve precise information instead of hallucinatin
Who is it for?
Developers building coding agents that must navigate large monorepos and documentation graphs without exhaustive file grepping.
Skip if: Simple keyword searches or projects without a prebuilt knowledge graph in AgentDB.
When should I use this skill?
An agent needs to explore dependencies, upstream causes, or indirect relationships starting from a named entity in the project knowledge graph.
What you get
Ranked traversal paths, pruned causal-edge chains, and synthesized context bundles from AgentDB graph queries.
- Ranked traversal paths
- Synthesized context from graph walk
Files
KG Traverse
Perform pathfinder graph traversal starting from a seed entity. Expands outward through causal edges, scores paths by relevance, and prunes low-similarity branches.
When to use
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.
Steps
1. Seed -- call mcp__claude-flow__agentdb_hierarchical-recall to look up the target entity by name 2. Expand -- call mcp__claude-flow__agentdb_causal-edge to find all edges connected to the seed entity, then recursively expand outward to the specified depth (default: 3) 3. Score -- for each path, compute relevance: cumulative_score = product(edge_weight * keyword_similarity(query, node)) using mcp__claude-flow__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__claude-flow__embeddings_generate + manual cosine, but that's not required for step 3 to function. 4. Prune -- remove paths with cumulative score below 0.3 5. Rank -- sort remaining paths by cumulative score descending 6. Synthesize -- call mcp__claude-flow__agentdb_context-synthesize to combine the top paths into a coherent summary 7. Report -- display the top 10 paths with: path (entity chain), relation types, cumulative score, and synthesized context
CLI alternative
npx @claude-flow/cli@latest memory search --query "relations for ENTITY_NAME" --namespace knowledge-graphRelated skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Use kg-traverse for multi-hop graph exploration from a seed entity; use flat memory search when you only need direct keyword matches.
FAQ
What MCP tools does kg-traverse use?
kg-traverse calls AgentDB MCP tools: agentdb_hierarchical-recall, agentdb_causal-edge, agentdb_pattern-search, and agentdb_context-synthesize, plus Bash for orchestration.
When should developers use kg-traverse?
kg-traverse fits when you need to explore the knowledge graph from a specific entity—finding dependents, upstream causes, or indirect relationships—with depth-controlled pathfinder traversal.