MoAI-ADK is a Python framework for building agentic AI workflows. It enables developers to create autonomous agents that can perform complex tasks. Operations teams benefit from automated workflows. It connects to Claude for AI capabilities.
git clone https://github.com/modu-ai/moai-adk.githttps://adk.mo.ai.kr
["1. Identify the task you want to automate and break it down into smaller sub-tasks that can be handled by individual agents.","2. Determine which agents you need and what responsibilities each agent should have. Consider using Claude's capabilities for complex tasks that require natural language processing or decision making.","3. Write the Python code for your workflow using the MoAI-ADK framework. Make sure to include the necessary imports and initialize the Claude client with your API key.","4. Define your agents as classes that inherit from the Agent class. Implement the run method for each agent, which should contain the logic for performing the agent's responsibilities.","5. Define your workflow as a class that inherits from the Workflow class. Include your agents in the workflow and implement the run method, which should contain the logic for running the agents in sequence or parallel.","Tip: Use the MoAI-ADK documentation and examples as a reference when writing your code. Make sure to test your workflow thoroughly before deploying it to production."]
Automate the generation of project specifications to streamline the development process.
Integrate AI agents to assist in code refactoring while preserving existing functionality.
Utilize LSP integration for real-time diagnostics and feedback during the coding workflow.
Implement parallel development environments to enhance team collaboration on large projects.
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/modu-ai/moai-adkCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Create a MoAI-ADK workflow that automates [TASK] for our operations team. The workflow should include [NUMBER] agents with the following responsibilities: [AGENT RESPONSIBILITIES]. Each agent should use Claude's capabilities to [SPECIFIC AI TASKS]. Provide the Python code for this workflow and explain how it integrates with our existing systems.
Here's a MoAI-ADK workflow that automates incident response for our operations team. This workflow includes 3 agents with the following responsibilities:
1. Incident Detection Agent: Monitors our systems for anomalies and creates incident tickets when issues are detected.
2. Incident Analysis Agent: Uses Claude to analyze incident data and determine the root cause.
3. Incident Resolution Agent: Implements predefined remediation steps and escalates to human operators when necessary.
```python
# Import necessary modules
from moai_adk import Agent, Workflow
from moai_adk.clients import ClaudeClient
# Initialize Claude client
claud_client = ClaudeClient(api_key='your_api_key_here')
# Define agents
class IncidentDetectionAgent(Agent):
def __init__(self):
super().__init__(name='Incident Detection Agent')
async def run(self):
# Monitor systems for anomalies
# Create incident tickets when issues are detected
pass
class IncidentAnalysisAgent(Agent):
def __init__(self):
super().__init__(name='Incident Analysis Agent')
self.claud_client = claud_client
async def run(self, incident_data):
# Use Claude to analyze incident data
# Determine the root cause
analysis = self.claud_client.analyze(incident_data)
return analysis
class IncidentResolutionAgent(Agent):
def __init__(self):
super().__init__(name='Incident Resolution Agent')
async def run(self, analysis):
# Implement remediation steps
# Escalate to human operators when necessary
pass
# Define workflow
class IncidentResponseWorkflow(Workflow):
def __init__(self):
super().__init__(name='Incident Response Workflow')
self.agents = [
IncidentDetectionAgent(),
IncidentAnalysisAgent(),
IncidentResolutionAgent()
]
async def run(self):
# Run agents in sequence
incident_data = await self.agents[0].run()
analysis = await self.agents[1].run(incident_data)
await self.agents[2].run(analysis)
# Run workflow
if __name__ == '__main__':
workflow = IncidentResponseWorkflow()
workflow.run()
```
This workflow integrates with our existing systems by using our incident management API to create and update tickets. The Incident Detection Agent monitors our systems using our existing monitoring tools. The Incident Resolution Agent uses our runbook automation system to implement remediation steps.Cloud ETL platform for non-technical data integration
IronCalc is a spreadsheet engine and ecosystem
Get more done every day with Microsoft Teams – powered by AI
Customer feedback management made simple
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan