ideon plan expand

Expand an existing series with new article ideas, backed by Google Keyword Planner research. The plan is presented in an interactive review flow before being saved to your queue.

Synopsis

ideon plan expand [series-slug] [options]

Arguments

Argument	Description	Required
series-slug	Series slug to expand	No (can be selected interactively)

If series-slug is omitted and not provided via --non-interactive, Ideon lists available series and prompts for selection interactively.

Options

Option	Alias	Description	Default
--publication	-p	Publication slug	Required
--country		Comma-separated ISO country codes	Publication default or US
--language		ISO 639-1 language code	Publication default or en
--article-count		Target new articles to plan	5
--seed-keywords		Comma-separated additional seed keywords	—
--content-type		Content type for queue entries	article
--model		Model for strong reasoning calls	deepseek/deepseek-v4-pro
--intent-model		Model for intent classification	deepseek/deepseek-v4-flash
--auto-save		Skip approval gates and save automatically	false
--non-interactive		Agent mode: plain text output to stdout	false
--dry-run		Run research but skip all writes	false

Examples

Basic expansion

ideon plan expand ai-deep-dives --publication tech-blog

If the series slug is omitted, an interactive prompt shows available series to pick from.

With custom article count

ideon plan expand kubernetes-series \
  --publication tech-blog \
  --article-count 8

Adding seed keywords

ideon plan expand cloud-cost \
  --publication finops-blog \
  --seed-keywords "cloud repatriation,AWS savings plans,reserved instances pricing" \
  --article-count 4

These additional keywords supplement the series's existing keywords for GKP research.

Non-interactive agent mode

ideon plan expand ai-deep-dives \
  --publication tech-blog \
  --non-interactive \
  --auto-save

Output goes to stdout. The plan is automatically persisted.

Dry-run to preview

ideon plan expand ai-deep-dives \
  --publication tech-blog \
  --dry-run

Runs research but persists nothing. Useful for scoping an expansion before committing.

How Expand Differs from Explore

Aspect	Explore (new-idea)	Expand (expand-series)
Starting point	Content idea from scratch	Existing series
Seed keywords	LLM-generated + user-provided	Series keywords + user-provided
Series output	Creates new series clusters	Plans articles for one existing series
Cluster formation	Groups candidates into new series	Uses the target series's structure
Coverage check	Full dedup against existing content	Dedup within the series scope
Queue entries	Articles queued under new series	Articles queued under existing series

Pipeline Stages

The expand mode skips clustering (since you're expanding a known series) and runs:

1. Hydrate — Load publication, series, output history, and GKP cache
2. Seeds — Extract keywords from the target series; apply seed keywords
3. Research — Iterative GKP queries
4. Score — KOB scoring, intent classification, candidate filtering
5. Plan Articles — Plan new articles for the existing series
6. Persist — Update series keywords and queue new articles

Interactive Flow

When --non-interactive is not set and --auto-save is not enabled:

1. Series selection (if series-slug was not provided) — Pick from available series
2. Plan review — Article details with keyword, intent, and format
3. Approval gate — Confirm or reject the plan

Exit Codes

Code	Meaning
0	Plan completed successfully
1	Pipeline failed (API error, missing credentials, series not found)
2	No results found

Output Format (Non-Interactive)

When --non-interactive is set, the output shows:

# Plan: expand
Mode: expand-series
Publication: tech-blog
Series: AI Deep Dives

## Research
Rounds: 2
Candidates evaluated: 45
Candidates passed: 18
Cache hits: 28
API calls: 5

## Articles

### Article: How Transformer Models Revolutionized NLP
Primary keyword: transformer models explained
Secondary keywords: attention mechanism, self-attention tutorial
Intent: informational
Funnel: top
Format: tutorial
Priority: high
Type: new

### Article: Transformers vs RNNs: A Practical Comparison
Primary keyword: transformers vs RNNs
Secondary keywords: sequential models comparison, LSTM alternatives
Intent: commercial
Funnel: middle
Format: comparison
Priority: medium
Type: new

Related Commands

• ideon plan explore — Research a new content idea
• ideon series add — Create a series manually
• ideon gkp ideas — Generate GKP keyword ideas
• ideon queue add — Add an article to the queue