Skip to main content

Quickstart

This guide gets you from zero to your first multi-output generation run.

1. Configure Settings and Secrets​

ideon settings

Inside settings, configure:

  • LLM model and model settings
  • T2I model and optional input overrides
  • Output directories
  • OpenRouter API key
  • Replicate API token

2. Generate Content Outputs​

ideon write "How small editorial teams can productionize AI writing" --primary article=1 --secondary x-thread=1 --secondary x-post=1 --style professional

Expected stages for article primary:

  1. Planning Shared Brief
  2. Planning Primary Article
  3. Writing Sections
  4. Expanding Image Prompts
  5. Rendering Images
  6. Generating Channel Content
  7. Enriching Links

If primary is not article, Ideon uses Planning Primary Content and Generating Primary Content, renders one primary cover image, and then generates secondary outputs.

3. Check Outputs​

By default (resolved from current working directory):

  • Generation directories: output/<timestamp>-<slug>/
  • Markdown files: article-1.md, x-thread-1.md, x-post-1.md, ...
  • Run metadata: job.json
  • Run analytics: generation.analytics.json
  • Shared generation assets: image files in the same generation directory

You can open the latest generation in browser preview:

ideon preview

Preview includes generation-level browsing, content-type tabs, and per-variant subtabs.

4. Run a Safe Dry Run​

ideon write --dry-run "How AI changes developer docs workflows"

Dry run keeps the full pipeline flow but skips OpenRouter and Replicate API calls while still producing generation artifacts.

5. Run with a Job File​

ideon write --job ./job.json

See Job Files for full schema and examples.

Next Steps​