Skip to content

Akordi Agents SDK

Build intelligent agents with LangGraph-enhanced workflows, guardrails, and multi-agent orchestration.

Overview

Akordi Agents SDK is a Python framework for building production-ready AI agents with AWS Bedrock. It provides:

  • Intelligent Agents - Create agents with built-in validation, tool use, and LLM integration
  • LangGraph Workflows - Advanced state management and conditional routing
  • Guardrails - AWS Bedrock guardrails integration for safe AI
  • Tool Orchestration - Intelligent tool selection and execution
  • Multi-Agent Systems - Coordinate multiple specialized agents
  • Chat History - DynamoDB-backed conversation persistence
  • Token Tracking - Monitor and analyze LLM usage

Minimal Working Example

This is the simplest way to create a working agent:

import os
from akordi_agents.core import create_langgraph_agent
from akordi_agents.services import AWSBedrockService

# Required: Set AWS region
os.environ["AWS_REGION"] = "us-east-1"

# Step 1: Create LLM service
llm_service = AWSBedrockService()

# Step 2: Create agent
agent = create_langgraph_agent(
    name="my_agent",
    llm_service=llm_service,
)

# Step 3: Process request
response = agent.process_request({
    "query": "What is machine learning?",
    "system_message": "You are a helpful assistant.",
    "max_tokens": 1000,
})

# Step 4: Get response
if response.get("success"):
    print(response["llm_response"]["response"])

Agent with Tools Example

Add custom tools to extend agent capabilities:

import os
from akordi_agents.core import create_langgraph_agent
from akordi_agents.services import AWSBedrockService
from akordi_agents.tools import Tool

os.environ["AWS_REGION"] = "us-east-1"

# Define a custom tool
class WeatherTool(Tool):
    def get_name(self) -> str:
        return "get_weather"

    def get_description(self) -> str:
        return "Get current weather for a location"

    def get_input_schema(self) -> dict:
        return {
            "type": "object",
            "properties": {
                "location": {"type": "string", "description": "City name"}
            },
            "required": ["location"]
        }

    def execute(self, **kwargs) -> dict:
        location = kwargs.get("location", "Unknown")
        return {"location": location, "temperature": "22°C", "conditions": "Sunny"}

# Create agent with tool
agent = create_langgraph_agent(
    name="weather_agent",
    llm_service=AWSBedrockService(),
    tools=[WeatherTool()],
    config={"enable_tools": True, "temperature": 0.1}
)

# Agent will automatically use tool when needed
response = agent.process_request({
    "query": "What's the weather in London?",
    "system_message": "You are a weather assistant. Use the weather tool to get current conditions.",
})

Features at a Glance

Installation

pip install akordi-agents

Or with Poetry:

poetry add akordi-agents

:octicons-arrow-right-24: Full Installation Guide

Architecture

graph TB
    subgraph "Akordi Agents SDK"
        A[AgentBuilder] --> B[CustomAgent]
        A --> C[LangGraphAgent]

        C --> D[ToolUseWorkflow]
        C --> E[MultiAgentWorkflow]

        D --> F[ValidationNode]
        D --> G[ToolDecisionNode]
        D --> H[ToolExecutionNode]
        D --> I[ResponseNode]

        B --> J[LLM Service]
        B --> K[Search Handler]
        B --> L[Validator]

        J --> M[AWS Bedrock]
        K --> N[Knowledge Base]
    end

    O[User Request] --> A
    I --> P[Response]

What's Next?

Requirements

  • Python 3.9+
  • AWS credentials (for Bedrock integration)
  • Poetry (recommended) or pip

License

MIT License - see LICENSE for details.