Skip to main content

Ideon Documentation

Ideon helps you convert a raw idea into a complete generation run with channel-specific outputs and optional generated imagery.

What Ideon Does​

  • Plans article structure and editorial angle when article output is requested
  • Drafts intro, sections, and conclusion for article outputs
  • Generates non-article outputs (X/LinkedIn/Reddit/newsletter/landing/blog) via channel-native prompts
  • Applies a run-level writing style overlay across all outputs
  • Expands image descriptions and renders cover/inline images for article-enabled generation runs
  • Writes generation artifacts (markdown files, job.json, generation.analytics.json, shared assets)

Who This Documentation Is For​

  • Operators and writers who want to run Ideon quickly
  • Engineers integrating Ideon into scripted or CI workflows
  • Contributors extending pipeline stages, models, and docs

Required Services for Live Runs​

  • OpenRouter API key
  • Replicate API token

If you only want to test end-to-end orchestration, use --dry-run to avoid external API calls.