Skip to content

AgentVault Examples

This section provides practical examples demonstrating how to use AgentVault and its components to build various multi-agent systems and integrations.

Core Concepts Examples

These examples illustrate specific features of the AgentVault client library (agentvault) and the server SDK (agentvault-server-sdk).

  • Basic A2A Server: Shows the minimal setup for creating an A2A-compliant agent using FastAPI and the Server SDK. Demonstrates handling tasks/send, get, cancel, and subscribe.
  • Direct Library Usage: Illustrates how to use the AgentVaultClient directly in Python to interact with an A2A agent, including task initiation and SSE event streaming.
  • OAuth2 Authenticated Agent: Demonstrates building an agent server that requires OAuth2 Client Credentials flow for authentication, including a mock token endpoint and protecting the A2A endpoint.
  • Stateful Agent: Shows how to build an agent that maintains state across multiple interactions within a single task lifecycle using the SDK's task store concepts.
  • Simple Communication Agents: Covers agents like the Task Logger, Registry Query (LLM Test Mode), and Simple Summary Agent, showcasing basic database interaction, LLM integration, and simple agent-to-agent calls.

Integration Examples

  • LangChain Tool Integration: Provides a template for wrapping an AgentVault A2A agent as a custom tool within the LangChain framework.

End-to-End Pipeline POCs

These Proof-of-Concept (POC) pipelines demonstrate how multiple specialized agents can collaborate to solve more complex problems using AgentVault orchestration principles.

  • SecOps Pipeline (LLM-Enhanced): 🔥 FEATURED 🔥 An advanced Security Operations pipeline using the Qwen3-8B LLM for real-time security alert analysis, investigation, and automated response. Demonstrates direct LLM integration, transparent reasoning, and interactive dashboard visualization. (Uses LangGraph for orchestration with Redis for real-time updates).

  • Research Pipeline: A sophisticated pipeline orchestrating 7 agents (topic research, crawling, extraction, verification, synthesis, editing, visualization) to generate comprehensive research reports on a given topic. Demonstrates complex workflow, artifact passing, and local artifact storage. (Uses LangGraph for orchestration).

  • Support Ticket Pipeline: Orchestrates 4 agents (ticket analysis, KB search, customer history, response suggestion) to process customer support tickets and suggest responses. Showcases integrating different data sources. (Uses LangGraph for orchestration).
  • Dynamics 365 Pipeline: A pipeline simulating integration with Dynamics 365 data. It involves fetching data, enrichment, health analysis, action recommendation, briefing generation, and task creation/notification execution. Demonstrates rule-based analysis, LLM usage, and executing actions based on insights. (Uses LangGraph for orchestration).
  • E-commerce Pipeline: Orchestrates agents for user profiling, product catalog lookup, trend analysis, and recommendation generation to provide personalized e-commerce suggestions. (Uses LangGraph for orchestration).
  • ETL Pipeline: Demonstrates an Extract, Transform, Load workflow using multiple agents and a database for artifact storage between steps. (Uses LangGraph for orchestration).
  • MCP Test Pipeline: Showcases the Model Context Protocol (MCP) by using a proxy agent to interact with MCP-compliant tool servers (filesystem, code runner) for executing specific operations. (Uses LangGraph for orchestration).