Vector Cluster

Groups code files by semantic similarity using graph-based community detection (spectral clustering or Louvain) via ruvector's graph hooks. You point it at a collection of files and it identifies natural clusters based on how the code relates. Useful when you're trying to understand the architecture of an unfamiliar codebase or find related functionality scattered across a project. Fair warning: the top-level cluster command in ruvector 0.2.25 is reserved for distributed operations that aren't ready yet, so you're working with the graph-cluster and graph-mincut hooks specifically. Outputs JSON with cluster assignments and cohesion scores that help you spot tight groupings versus loose associations.

skills.sh installs: 729
GitHub stars: 81.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add ruvnet/ruflo --skill vector-cluster --agent claude-code

Installs into .claude/skills of the current project.

Files

Vector Cluster

Cluster vectors in a namespace by semantic similarity using ruvector.

When to use

Use this skill when you have a collection of embeddings and want to discover natural groupings. Clustering reveals themes, identifies outliers, and helps organize large vector collections.

Steps

Ensure ruvector@0.3.13 is available:

npm ls ruvector 2>/dev/null | grep '0.3.3' || npm install ruvector@0.3.13

Run clustering — in ruvector@0.3.13 the only working clustering is via hooks graph-cluster (spectral/Louvain over a code graph). The top-level cluster command is reserved for distributed cluster ops and is currently "Coming Soon" upstream.
```
npx -y ruvector@0.3.13 hooks graph-cluster <files...>
npx -y ruvector@0.3.13 hooks graph-mincut <files...>
```
Review output — JSON with cluster assignments, community labels, and edges. If you see "graph.nodes is not iterable", run hooks init first to seed the graph state.
Store results: mcp__plugin_ruflo-core_ruflo__memory_store({ key: "clusters-PROJECT-TIMESTAMP", value: "CLUSTER_ASSIGNMENTS", namespace: "vector-clusters" })