๐ง AI Agents for Workflow AutomationMarch 22, 2026โ
Tests passing
Workflow Agent Builder
A library for easily creating AI-powered workflow automation agents using YAML-based task definitions. This tool allows developers to define workflows with natural language prompts and actions, which are interpreted and executed by AI agents.
What It Does
- YAML-based declarative workflow definitions
- Pluggable AI models (e.g., OpenAI)
- Support for conditional task branching and loops
Installation
Install the required dependencies:
pip install -r requirements.txtUsage
To execute a workflow:
python workflow_agent_builder.py --workflow tasks.yaml --api-key YOUR_API_KEYOptionally, you can save the output to a JSON file:
python workflow_agent_builder.py --workflow tasks.yaml --api-key YOUR_API_KEY --output results.jsonExample YAML Workflow
tasks:
- name: "Task 1"
prompt: "What is 2+2?"
max_tokens: 10
- name: "Conditional Task"
prompt: "What is the capital of France?"
condition: "Is this a test?"
max_tokens: 10Source Code
import yaml
import json
import click
import openai
from typing import Any, Dict, List
class WorkflowAgentBuilder:
def __init__(self, model: str = "text-davinci-003", api_key: str = None):
self.model = model
if api_key:
openai.api_key = api_key
def execute_task(self, task: Dict[str, Any]) -> Any:
"""Executes a single task using the AI model."""
prompt = task.get("prompt", "")
if not prompt:
raise ValueError("Task is missing a 'prompt' field.")
try:
response = openai.Completion.create(
engine=self.model,
prompt=prompt,
max_tokens=task.get("max_tokens", 100),
temperature=task.get("temperature", 0.7)
)
return response.choices[0].text.strip()
except Exception as e:
return f"Error executing task: {str(e)}"
def execute_workflow(self, workflow: Dict[str, Any]) -> List[Dict[str, Any]]:
"""Executes a workflow defined in a YAML file."""
results = []
tasks = workflow.get("tasks", [])
for task in tasks:
if "condition" in task:
condition_result = self.execute_task({"prompt": task["condition"]})
if condition_result.lower() != "true":
continue
result = self.execute_task(task)
results.append({"task": task.get("name", "Unnamed Task"), "result": result})
return results
@click.command()
@click.option('--workflow', type=click.Path(exists=True), required=True, help="Path to the YAML workflow file.")
@click.option('--output', type=click.Path(), required=False, help="Path to save the JSON output.")
@click.option('--api-key', type=str, required=True, help="API key for the AI model.")
def main(workflow, output, api_key):
"""CLI entry point for the Workflow Agent Builder tool."""
try:
with open(workflow, 'r') as file:
workflow_data = yaml.safe_load(file)
builder = WorkflowAgentBuilder(api_key=api_key)
results = builder.execute_workflow(workflow_data)
if output:
with open(output, 'w') as outfile:
json.dump(results, outfile, indent=4)
else:
click.echo(json.dumps(results, indent=4))
except Exception as e:
click.echo(f"Error: {str(e)}", err=True)
if __name__ == "__main__":
main()
Community
Downloads
ยทยทยท
Rate this tool
No ratings yet โ be the first!
Details
- Tool Name
- workflow_agent_builder
- Category
- AI Agents for Workflow Automation
- Generated
- March 22, 2026
- Tests
- Passing โ
Quick Install
Clone just this tool:
git clone --depth 1 --filter=blob:none --sparse \ https://github.com/ptulin/autoaiforge.git cd autoaiforge git sparse-checkout set generated_tools/2026-03-22/workflow_agent_builder cd generated_tools/2026-03-22/workflow_agent_builder pip install -r requirements.txt 2>/dev/null || true python workflow_agent_builder.py