Intelligence Layer Skill

RAG with knowledge graph and vector search combines Neo4j for relationship-based queries and Qdrant for semantic search, enabling retrieval-augmented generation through multi-agent orchestration. This Skill deploys both layers together with CrewAI and Temporal workflows for end-to-end reasoning and retrieval.

Yes. Neo4j stores relational knowledge graphs while Qdrant performs vector-based semantic search. This Skill integrates both with Docker Compose, AnythingLLM RAG hooks, and CrewAI agents to enable graph-based reasoning combined with semantic retrieval in a single deployment.

Multi-agent orchestration with CrewAI combined with Temporal workflow engine provides durable, stateful task coordination. This Skill deploys CrewAI for agent collaboration and Temporal for long-running workflows, ensuring reliable execution and state persistence across distributed reasoning tasks.

This Skill deploys a complete Docker Compose stack including Neo4j with APOC plugins, Qdrant vector store, Temporal workflow engine, and AnythingLLM-based RAG, providing immediately operational infrastructure for LLM-powered graph reasoning and semantic search.

This Skill unifies knowledge graphs, vector search, multi-agent coordination, and RAG in one integrated deployment. Neo4j, Qdrant, CrewAI, and Temporal work together through defined integration interfaces, eliminating the need to stitch separate tools for graph reasoning, semantic retrieval, and workflow execution.

Combining multiple systems requires careful data synchronization between Neo4j and Qdrant, agent-level error handling in CrewAI, and Temporal workflow observability. This Skill provides the deployment foundation, but production use demands monitoring vector-to-graph consistency, multi-agent failure recovery, and workflow auditing.