Stage 1: Local Agent Execution

Your on-ramp to CAF. No federation required.

Task

Implement and run a simple, persistent, stateful agentic application on a laptop or workstation, using local or remote LLMs.

Why This Matters

Local execution lets you develop and test agent logic before deploying to HPC. LangGraph and Academy specifications are reproducible and portable—the same agent definition runs locally or at scale.

Details

Aspect Value
Technologies LangGraph, Academy
Where code runs Laptop, workstation, VM
Scale Single agent / small multi-agent
Status Mature

Getting Started

Building Scientific Agents — A three-stage guide covering:

  1. LLM Agents — Build agents that reason and call tools (LangGraph)
  2. Distributed Agents — Run agents across machines (Academy)
  3. Production Agents — Combine LangGraph + Academy for real deployments
Additional Guides Description
LLM Configuration Configure OpenAI, Ollama, or FIRST backends

Architecture

Here we deal with agentic applications in which one or more agents operate entirely locally, each potentially calling tools and/or LLMs.

Local agent architecture: Agent calls Tools and LLM

Examples

Minimal Example: Calculator Agent

The simplest possible agent—an LLM that can use a calculator tool:

Example Framework Code
AgentsCalculator LangChain + LangGraph View 
@tool
def calculate(expression: str) -> str:
    """Evaluate a mathematical expression."""
    return str(eval(expression, {"__builtins__": {}}, {}))

llm = ChatOpenAI(model="gpt-4o-mini")
agent = create_react_agent(llm, [calculate])
agent.invoke({"messages": [HumanMessage(content="What is 347 * 892?")]})

Tool Pattern Examples

These examples demonstrate different types of tools an agent can use:

Example Tool Type What It Shows
AgentsRAG Vector search Retrieval-augmented generation from scientific documents
AgentsDatabase Data queries Natural language queries on pandas DataFrames
AgentsAPI External APIs Calling PubChem for chemical information
AgentsConversation Memory Stateful conversations with short and long-term memory

Each example follows the same pattern as the Calculator but with more realistic, science-relevant tools.

Academy Examples

These examples demonstrate Academy framework patterns for distributed agent coordination:

Example Pattern Description
AgentsAcademyBasic Basics Two agents communicating - the “Hello World” of Academy
AgentsRemoteTools Remote Tools Coordinator calls tools on a ToolProvider agent
AgentsHybrid Hybrid Academy + LangGraph: distributed agents with LLM reasoning
AgentsPersistent Persistence Checkpoint and resume workflows across restarts
AgentsFederated Federation Cross-institutional collaboration (ANL, ORNL, LBNL)

Five-Agent Scientific Discovery Pipeline

This more involved example demonstrates multi-agent coordination for scientific workflows. Five specialized agents work in sequence, each contributing domain expertise before passing results to the next:

Agent Role Input Output
Scout Surveys problem space, detects anomalies Goal Research opportunities
Planner Designs workflows, allocates resources Opportunities Workflow plan
Operator Executes the planned workflow safely Plan Execution results
Analyst Summarizes findings, quantifies uncertainty Results Analysis summary
Archivist Documents everything for reproducibility Summary Documented provenance

5-agent pipeline: Scout → Planner → Operator → Analyst → Archivist

Implementations of Five-Agent Workflow

We provide implementations of this example in LangGraph and Academy, demonstrating different orchestration patterns. Note that these implementations are toys: they create agents that communicate, but each agent’s internal logic is just a stub.

Example Framework Pattern Code
AgentsLangGraph LangGraph Graph-based orchestration View
AgentsAcademy Academy True pipeline (agent-to-agent) View
AgentsAcademyHubSpoke Academy Hub-and-spoke (main orchestrates) View

Pattern comparison:

Hub-and-Spoke vs Pipeline patterns

No LLM is used in the Academy examples—agent logic is stubbed to focus on the messaging patterns.

Dashboard Version

The dashboard version wraps agents with a full-screen Rich UI showing live progress across multiple scientific goals.

Example Framework Features Code
AgentsAcademyDashboard Academy Rich dashboard, multi-goal View