logo
AgentLed

agentled create

Create workflows from natural language descriptions. The CLI analyzes your prompt, selects integrations, configures steps, and builds a ready-to-run workflow.

Syntax

agentled create "<description of what the workflow should do>"

The AI interprets your description, picks the right integrations from 100+ available apps, configures each step with the correct parameters, and scaffolds a complete workflow. The result is saved as a draft you can review, edit, and publish.

Examples

Outbound Campaign

agentled create "Find fintech CTOs in Europe, enrich via
LinkedIn + Hunter, score by ICP fit, draft personalized outreach"

Research Agent

agentled create "Research competitors in the AI orchestration
space, scrape their pricing pages, and summarize in a table"

Content Pipeline

agentled create "Scrape trending AI topics, generate 5 LinkedIn
posts with different angles, and schedule them for the week"

Monitoring

agentled create "Monitor Hacker News for mentions of our brand,
score sentiment, alert on Slack if negative"

How It Works

1.Parse intent. The AI analyzes your description to identify the goal, target data, and desired outputs.

2.Select integrations. It picks the right apps (Hunter, LinkedIn, CRM, web scraping, AI models) based on what the workflow needs.

3.Configure steps. Each step is set up with the correct action, input mappings, and AI prompts.

4.Save as draft. The workflow is created in draft status so you can review and refine before publishing.

From Create to Run

After creating a workflow, use the CLI to run and manage it:

# Create the workflow
agentled create "Outbound to fintech CTOs in Europe"

# List workflows to find the ID
agentled workflows list

# Publish the draft
agentled workflows publish <id> --status live

# Run it
agentled workflows start <id>

# Check execution status
agentled executions list <id>

Using via MCP

You can also create workflows through the MCP chat tool from any connected AI client:

> "Create a workflow that finds SaaS CTOs on LinkedIn,
   enriches emails via Hunter, scores by ICP,
   and saves to Knowledge Graph"

The MCP chat and create_workflow tools provide the same AI-powered creation flow as the CLI.

Tips

  • •Be specific. “Find fintech CTOs in Europe” produces better results than “find leads.”
  • •Mention integrations. If you want a specific tool (Hunter, LinkedIn, Slack), name it in the description.
  • •Describe the output. Say what you want at the end: “save to Knowledge Graph,” “send via email,” “post to Slack.”
  • •Review the draft. AI-generated workflows are a starting point. Check the steps, adjust prompts, and test before going live.

Next Steps