Workflows

Sequential caching, parallel execution, and conditional branching patterns.

Workflows that actually orchestrate - not just sequential agent calls. Each example demonstrates real workflow value: phase caching, parallel coordination, or conditional routing.

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.workflow import Workflow, Step

researcher = Agent(name="Researcher", model=OpenAIResponses(id="gpt-5.2"))
writer = Agent(name="Writer", model=OpenAIResponses(id="gpt-5.2"))

workflow = Workflow(
    name="Content Pipeline",
    steps=[
        Step(agent=researcher, task="Research the topic"),
        Step(agent=writer, task="Write based on research"),
    ],
)

workflow.run("AI trends in 2025")

Examples

Blog Post Generator

Three-phase pipeline with session caching. Research → Extract → Write.

Company Description

Parallel data gathering from 4 sources, then synthesis.

Employee Recruiter

Conditional branching: only interviews candidates above score threshold.

Notion Manager

Classification → routing to create/update/search Notion pages.

Workflow Patterns

Pattern	Example	Description
Sequential + Caching	Blog Post Generator	Phase checkpoints for efficiency
Parallel → Sequential	Company Description	Gather concurrently, then synthesize
Conditional Branching	Employee Recruiter	Different paths based on output
Classification Routing	Notion Manager	Route to different actions by type

See Workflow Patterns for detailed documentation.