Memory Search

This is a seriously full-featured semantic search system that goes way beyond basic vector similarity. You get three retrieval modes: standard dense embeddings for speed, hybrid sparse+dense fusion that benchmarks 20-49% better on mixed queries, and graph RAG for multi-hop reasoning that pulls 30-60% gains on complex questions. The smart mode runs a five-phase pipeline with query expansion, MMR diversity filtering, and recency decay weighting. It searches across namespaced memory stores (patterns, tasks, solutions, feedback, security) and can synthesize context from multiple sources. If you're building anything with conversational memory or need to retrieve related information across sessions, the architecture here is genuinely sophisticated.

skills.sh installs: 751
GitHub stars: 81.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add ruvnet/ruflo --skill memory-search --agent claude-code

Installs into .claude/skills of the current project.

Files

Memory Search

Search Ruflo's stored memories. Select a strategy supported by the Ruflo CLI or MCP tool, and report which one actually ran.

Strategy selection

Default: semantic retrieval through memory search or MCP memory_search.
--hybrid: CLI keyword + semantic result union, deduplicated and sorted by score. This is not the ruvector search CLI and does not claim Graph RAG or sparse/dense RRF.
--smart: request query expansion, RRF across variants, recency weighting, and MMR diversity. The MCP tool can fall back to plain search if the installed memory package lacks smartSearch; inspect the returned fallback note before claiming SmartRetrieval ran.
Cross-namespace: MCP memory_search_unified merges and deduplicates results from multiple namespaces.

If the user requests --graph-rag, explain that this plugin has no supported text-query Graph RAG command. Offer --smart semantic retrieval as a clearly labeled fallback; do not claim graph traversal.