OmniCoreAgent Cookbook

Copy, paste, run. These examples show how to build with OmniCoreAgent from the first local agent through production-shaped application harnesses.

🚀 Where to Start

New to OmniCoreAgent? Start with getting_started — progressive examples from “Hello World” to production deployments. Know what you’re building? Find your use case below. Just want to see it work?

from omnicoreagent import OmniCoreAgent

agent = OmniCoreAgent(
    name="my_agent",
    system_instruction="You are a helpful assistant.",
    model_config={"provider": "openai", "model": "gpt-5.4"}
)

result = await agent.run("Hello!")
print(result["response"])

📚 Build by Use Case

🤖 I want to build my first agent

→ getting_started — Basic agents, tools, memory, events, and configuration.

🛠️ I want to use custom tools

→ getting_started/agent_with_local_tools.py — Register Python functions as tools.

🔌 I want to connect MCP servers

→ getting_started/agent_with_mcp_tools.py — Connect to external MCP servers.

🚁 I want agents running autonomously

→ background_agents — Scheduled tasks and queue-based execution.

🧠 I want complex multi-step workflows

→ getting_started/agent_with_sub_agents.py — OmniCoreAgent dynamic subagents, workspace files, and orchestration.

🏭 I want to build a real application

→ real_applications — Production-shaped app harness examples that show domain tools, workspace files, offloading, telemetry, guardrails, and workspace command tools across different apps.

🧪 I want more domain demos

→ advanced_agent — E-commerce, flight booking, and customer support.

🌐 I want to deploy as an API

→ omniserve — Deploy agents as REST/SSE APIs with one command.

🚀 I want to run in production

→ production — Metrics and guardrails.

📂 Directory Structure

cookbook/
├── getting_started/     # Start here! Progressive learning path
├── background_agents/   # Scheduled and autonomous agents
├── omniserve/           # 🆕 Deploy agents as REST/SSE APIs
├── real_applications/   # End-to-end application harness examples
├── advanced_agent/      # Real-world application examples
└── production/          # Production-ready configurations

🎯 Featured Examples

Example	What You’ll Build
Research Due Diligence	Research agent with parallel tools, workspace reports, artifact readback, telemetry
Support Operations	Support agent with customer/order tools, guardrails, escalation, workspace notes
Workspace Code Review	Code review agent using built-in workspace file commands instead of unrestricted shell
Support Background Task	Real support operations app running through durable background task state, events, attempts, and workspace output
E-commerce Shopper	Personal shopping assistant with cart, preferences, recommendations
Flight Booking	Travel agent with search, booking, and itinerary management
Customer Support	Support agent with ticket handling and escalation

🤝 Contributing

Want to add a cookbook example? See CONTRIBUTING.mdx for guidelines.

​OmniCoreAgent Cookbook

​🚀 Where to Start

​📚 Build by Use Case

​🤖 I want to build my first agent

​🛠️ I want to use custom tools

​🔌 I want to connect MCP servers

​🚁 I want agents running autonomously

​🧠 I want complex multi-step workflows

​🏭 I want to build a real application

​🧪 I want more domain demos

​🌐 I want to deploy as an API

​🚀 I want to run in production

​📂 Directory Structure

​🎯 Featured Examples

​🤝 Contributing