Knowledge Management

Centralized Knowledge Bases

Store, version, and govern knowledge embeddings in one place. Share context across agents while maintaining access control and compliance.

Vector Store Management

Manage multiple knowledge bases with versioning and access control. Connect any vector database while centralizing governance in Eevee.

  • Support for Pinecone, Weaviate, Qdrant, Chroma
  • Automatic embedding versioning
  • Cross-agent knowledge sharing

Intelligent Retrieval

Advanced RAG capabilities with hybrid search, reranking, and context optimization. Ensure agents get the most relevant knowledge for every query.

  • Hybrid semantic + keyword search
  • Automatic reranking and filtering
  • Context window optimization

Access Control

Fine-grained permissions for knowledge bases. Control which agents and users can access sensitive documents and data.

  • Role-based access control (RBAC)
  • Per-document access policies
  • Audit logs for compliance

Dynamic Updates

Add, update, or remove documents without redeploying agents. Knowledge bases update in real-time across all applications.

  • Hot reload for knowledge updates
  • Incremental indexing for large datasets
  • Automatic cache invalidation

Load knowledge on demand

Reference knowledge bases from Eevee Server with built-in access control

rag_agent.py
import eevee as ev

# Load knowledge base from Eevee Server
kb = ev.load("knowledge", "product_docs")

# Use with RAG agent
agent = Agent(
    tools=[kb.search],
    system_message="""
    Use the product_docs knowledge base
    to answer customer questions.
    """
)

# Knowledge updates automatically
# Access control enforced by Eevee policies

Centralize your knowledge

Stop managing embeddings across silos - govern knowledge in one place