Skip to content
Agent

ADK Multi-Agent Architecture

by kubiyabot

AI Summary

ADK Multi-Agent Architecture enables developers to build serverless AI agents with function calling capabilities integrated directly into workflows, ideal for DevOps, SRE, and automation teams needing intelligent decision-making at scale.

Install

Copy this and paste it into Claude Code, Cursor, or any AI assistant:

I want to set up the "ADK Multi-Agent Architecture" agent in my project.

Please run this command in my terminal:
# Add AGENTS.md to your project root
curl --retry 3 --retry-delay 2 --retry-all-errors -o AGENTS.md "https://raw.githubusercontent.com/kubiyabot/sdk-py/main/docs/kubiya/providers/adk/agents.mdx"

Then explain what the agent does and how to invoke it.

Description

Deep dive into ADK's serverless AI agents and function calling

Custom Tools

Add custom tools to agents: `python from kubiya.providers.adk.tools import Tool

Serverless AI Agents at Your Service

<Card title="🚀 Serverless Agents with Function Calling" icon="robot"> ADK exposes Kubiya's serverless AI agents that leverage state-of-the-art techniques: • Function Calling: Agents use simple function calls to execute workflow tools • Serverless Execution: Spin up on-demand, no persistent processes • Tool Integration: Seamlessly call workflow steps as functions • Best-in-Class Models: Leverages top AI models for reasoning • Deterministic Outputs: Structured responses that fit into workflows </Card>

The Power of Inline Agents

Unlike traditional agent frameworks that require complex setups, Kubiya's inline agents are: • Just workflow steps - No separate agent infrastructure • Function-based - Agents call tools through simple function interfaces • Context-aware - Access workflow state and previous outputs • Serverless - No always-on processes consuming resources • Integrated - Part of your workflow DAG, not external `python

Serverless agent as a workflow step

.step("intelligent-analysis") .inline_agent( message="Analyze system health and recommend actions", agent_name="sre-assistant", ai_instructions="You are an SRE expert. Use the provided tools to gather metrics.", # Agent uses function calling to execute tools tools=[ { "name": "get_cpu_usage", "type": "function", "description": "Get current CPU usage across pods", "implementation": "kubectl top pods | awk '{print $2}'" }, { "name": "check_error_logs", "type": "function", "description": "Count errors in application logs", "implementation": "kubectl logs -l app=myapp | grep ERROR | wc -l" } ], runners=["kubiya-hosted"] # Serverless execution ) .output("SRE_ANALYSIS") `

Discussion

0/2000
Loading comments...

Health Signals

MaintenanceCommitted 1mo ago
Active
AdoptionUnder 100 stars
2 ★ · Niche
DocsMissing or thin
Undocumented

GitHub Signals

Stars2
Issues17
Updated1mo ago
View on GitHub
No License

My Fox Den

Community Rating

Sign in to rate this booster

Works With

Claude Code
Claude.ai