Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@telepat/ideon) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

ooooo oooooooooo.   oooooooooooo   .oooooo.   ooooo      ooo
`888' `888'   `Y8b  `888'     `8  d8P'  `Y8b  `888b.     `8'
 888   888      888  888         888      888  8 `88b.    8
 888   888      888  888oooo8    888      888  8   `88b.  8
 888   888      888  888    "    888      888  8     `88b.8
 888   888     d88'  888       o `88b    d88'  8       `888
o888o o888bood8P'   o888ooooood8  `Y8bood8P'  o8o        `8

AI Writer Extraordinaire

Ideon is a TypeScript CLI that turns an idea into one or more Markdown outputs, with optional generated images for article runs.

Features

End-to-end pipeline with stage visibility: planning, sections, image prompts, image rendering, and output assembly
Interactive terminal UI with clear per-stage status and summaries
Non-interactive fallback logging for CI and piped runs
Resume support through local stage checkpoints in .ideon/write/state.json
Config precedence across saved settings, job files, environment variables, and CLI flags
Secure secret storage in OS keychain (OpenRouter + Replicate tokens)
Runtime validation for generated plan and image prompt payloads
Retry + timeout hardening on OpenRouter requests

Installation

Prerequisites:

Node.js 20+
npm 10+

Install globally:

npm i -g @telepat/ideon

Verify installation:

ideon --help

Getting Started

Configure credentials interactively:

ideon settings

Generate your first article:

ideon write "How small editorial teams can productionize AI writing"

Generate multi-output runs:

ideon write "How small editorial teams can productionize AI writing" --target article=1 --target x-post=2 --style professional

Run a safe pipeline dry run (no provider calls):

ideon write --dry-run "How AI changes technical publishing"

Core Commands

ideon settings
ideon write "An article idea"
ideon write --job ./job.json
ideon write --dry-run "An article idea"
ideon write resume
ideon delete my-article-slug
ideon preview

Preview Generated Articles

Serve the latest generated article locally with assets and open it in your browser:

ideon preview

This launches the new React preview app (served from dist/preview) and the preview API server.

You can also preview a specific article and choose a port:

ideon preview ./output/my-article.md --port 4173

If you are iterating on preview UI code in src/preview-app, rebuild client assets after UI changes:

npm run build:preview

Credentials

Live runs require:

IDEON_OPENROUTER_API_KEY
IDEON_REPLICATE_API_TOKEN

You can set these as environment variables, or save them via ideon settings (recommended).

Output

By default, Ideon writes:

Generation directories: /output/<timestamp>-<slug>/
Markdown outputs per target: article-1.md, x-1.md, linkedin-1.md, and others
Run artifacts per generation: job.json, generation.analytics.json
Local resume artifacts: .ideon/write/state.json

Development Scripts

npm run lint
npm test
npm run build
npm run preview
npm run pricing:refresh

Documentation

User and technical docs site source: docs-site/
Start docs locally: npm run docs:start
Build docs: npm run docs:build

Links:

GitHub repository: telepat-io/ideon
npm package: @telepat/ideon
Documentation site: docs.telepat.io/ideon

Key docs:

CLI commands: docs-site/docs/reference/cli-reference.md
Configuration and precedence: docs-site/docs/guides/configuration.md
Pipeline and resume: docs-site/docs/guides/pipeline-stages.md
Output artifacts: docs-site/docs/guides/output-structure.md
Performance tuning: docs-site/docs/guides/performance-and-costs.md

GitHub Pages URL:

https://docs.telepat.io/ideon