Skip to main content

One Idea. Endless Formats.

Ideon turns a single idea into a complete content campaign — publish-ready drafts across every channel, all sharing one voice, one style, and one strategy.

Built for marketers, founders, and lean teams who need to ship high-quality content at scale without hiring a full editorial department.

Write Once, Publish Everywhere​

Turn one idea into an article, blog post, newsletter, X thread, X post, LinkedIn post, and Reddit post — all in a single run. Your article is the anchor. Everything else promotes it.

ideon write "Our new AI feature" \
  --primary article=1 \
  --secondary x-thread=2 \
  --secondary linkedin-post=1 \
  --secondary reddit-post=1

No more manually rewriting the same announcement seven different ways. Ideon adapts structure, length, and tone for each channel while keeping the core message intact.

Writing That Sounds Like You​

Pick a style and an intent. Every output in the run shares the same voice — so your X thread sounds like it came from the same writer as your article.

StyleIntentResult
technicaltutorialEducational deep-dives developers actually read
professionalcase-studyCredibility-building founder content
storytellingannouncementProduct launches that hook from the first line
persuasiveopinion-pieceThought leadership with edge
friendlyhow-to-guideApproachable, shareable tutorials

13 styles. 13 intents. One consistent voice across every channel.

Publications and Series​

Organize your content strategy with publications and series.

Publications let you define editorial policies, default styles, intents, and audience hints per publication. Set it once, and every write run under that publication inherits the right voice.

Series group related articles under a shared topic and editorial thread. A series can override publication defaults, and every article written under it gets contextual prompt injection — the LLM knows it's part of a larger narrative and maintains thematic coherence.

# Create a publication
ideon publication add "Tech Blog" --style technical --intent tutorial --tone authoritative

# Create a series under it
ideon series add "AI Deep Dives" --topic "Exploring cutting-edge AI technologies" --publication tech-blog

# Write with series context
ideon write "How RAG systems work" --series ai-deep-dives --primary article=1
  • Layered defaults — saved settings → job → env → publication → series → CLI flags
  • Override anything — series can override style, intent, length, content targets, model settings, and editorial policy
  • Standalone or linked — series work with or without a publication
  • Thematic injection — series name and topic are injected into every prompt for coherent multi-article arcs

Authors and Editorial Checklist​

Give every draft a real voice. Authors are first-class profiles — name, slug, and a freeform profile (experience, credentials, writing style). When resolved for a run, author context is injected into all writing prompts (plan, sections, channel content).

# Create an author
ideon author add "Alex Chen" --profile "Staff SRE. Writes about Kubernetes and incident response."

# Set defaults on publication or series
ideon publication edit tech-blog --author alex-chen
ideon series edit ai-deep-dives --author alex-chen --experience "Standing anecdotes for this series"

# Override per run
ideon write "How we fixed prod" --author alex-chen \
  --experience "The readiness probe false positive from Q2" \
  --primary article=1
  • Resolution chain — run --author → series defaultAuthor → publication defaultAuthor
  • Experience notes — series standing notes plus per-run --experience, concatenated when both are present
  • Weave-when-provided — supplied anecdotes may appear in first person; without them, third-person expert voice or [AUTHOR: …] placeholders
  • Draft-first publishing — no bylines or AI disclosures in the draft body; humans add Who/How at publish time
  • Editorial checklist — dynamic pre-publish items written to meta.json and printed after each run (byline, AI disclosure, placeholder cleanup, stat verification)

Content Queue​

Plan your content pipeline ahead of time with the content queue. Add articles to a global queue with full parameter snapshots, then write them one at a time when you're ready.

# Queue articles for later
ideon queue add "How RAG systems work" --primary article=1 --publication tech-blog --style technical
ideon queue add "Our Q3 product launch" --primary article=1 --secondary x-thread=2 --intent announcement

# See what's queued
ideon queue list

# Write the next one
ideon write --from-queue

# Write the next one for a specific publication
ideon write --from-queue --publication tech-blog
  • Self-contained snapshots — publication and series defaults are resolved at enqueue time
  • Atomic dequeue — concurrent writes can't double-pick the same entry
  • Auto-revert — failed or interrupted writes restore the entry to the queue
  • Override at write time — CLI flags override queued settings

Data-Backed Content Planning​

Stop guessing what to write. Ideon's plan command researches your content idea against real Google Keyword Planner data, scores keyword opportunities, groups them into thematic series, and plans individual articles — all reviewed through an interactive terminal UI before hitting your queue.

# Explore a new topic
ideon plan explore "Content strategy for B2B SaaS" \
  --publication tech-blog \
  --series-count 3 \
  --articles-per-series 5

# Expand an existing series
ideon plan expand ai-deep-dives \
  --publication tech-blog \
  --article-count 6
  • Dual modes — explore for new topics, expand for growing existing series
  • GKP-powered — Real search volume, competition, and CPC data back every decision
  • KOB scoring — Keyword Opportunity Benchmark weights volume, intent, and competition to prioritize what matters
  • Topic clustering — LLM groups shortlisted keywords into coherent series with pillar keywords and funnel stages
  • Coverage-aware — Skips keywords you've already published; surfaces stale content for refresh
  • Interactive review — Navigate series and articles in the terminal; approve or reject before saving
  • Agent-ready — --non-interactive and --auto-save for CI and automation workflows

Learn more →

Research-Backed Drafts​

Ideon browses the web and inserts contextual external links like a human writer would — no manual research, no generic placeholder URLs. Just credible, relevant links that add depth and authority to your drafts.

SEO-Optimized From the Start​

Ideon's writing pipeline enforces on-page SEO best practices at every stage of content generation — not as an afterthought, but as a baked-in layer of the writing process.

During planning, titles are constrained to search-safe lengths, meta descriptions are shaped for click-through impact, and the planner assigns a primaryKeyword plus per-section targetKeywords (0–2 each) with parity placement rules for both user-provided and LLM-generated keyword sets. During writing, tiered guide bundles (intro / section / outro) and the keyword-integration guide shape placement — primary keyword in the title and intro, section targets in BLUF openers. After section writing, a default-on seo-check stage runs deterministic lint and, when needed, a five-tool surgical editor agent (errors-only pass mode by default; strict optional) that fixes prose and metadata without restructuring the article.

During writing, four dedicated SEO guides shape every long-form section:

  • On-page essentials — heading hierarchy, BLUF paragraphs, key takeaway blocks, and paragraph structure optimized for both human readers and search crawlers
  • E-E-A-T signals — weave supplied author experience only; competing viewpoints and primary-source citations; no invented first-person practitioner stories
  • Fact density — substantial value beyond the obvious; facts and citations when they genuinely support the section (soft length-scaled targets, not quotas)
  • AI search extraction — question-shaped headings for informational sections, self-contained paragraphs, comparison tables for multi-option sections, and anti-gaming rules for generative answer visibility

Optional FAQ block — For informational intents on long-form primary targets (tutorial, how-to-guide, cornerstone, deep-dive-analysis, case-study, roundup-curation), Ideon can append a ## FAQ section after the conclusion via a dedicated sections:faq LLM call. Control with faqSection in settings or --faq-section / --no-faq-section on the CLI. Dry-run mode generates placeholder FAQ content without API calls.

No keyword stuffing. No SEO hacks. Just disciplined writing that happens to perform in both traditional search results and generative AI summaries.

Your Model, Your Choice​

Plug in any LLM via OpenRouter. Switch models without changing your workflow. Use Claude for nuance, GPT-4 for speed, or any model OpenRouter supports — Ideon works the same way regardless.

Built on Proven Writing Principles​

Ideon's writing engine is grounded in a guide-first prompt composition system compiled from a large corpus of real writing advice and best practices. Every generation follows concrete rules for:

  • Structure and clarity
  • Specificity over vagueness
  • Rhythm and scannability
  • Active voice with concrete subjects
  • Channel-native delivery

No generic AI filler. No over-polished transitions. Just disciplined, human-sounding prose.

Smarter Than a Skill. Cheaper Than Raw Prompts.​

Most AI content tools treat the LLM like a black box you feed everything into. Ideon doesn't. It drives the entire pipeline with deterministic code — planning, orchestration, formatting, file management, and state tracking — and only calls the model when actual language generation is needed.

You pay for tokens when they matter (drafting prose, planning structure, expanding ideas) and never for things code does better (parsing arguments, routing outputs, managing runs, rendering Markdown). Compared to agent skills or general-purpose LLM workflows that burn context windows on orchestration chatter, Ideon is leaner, faster, and dramatically cheaper at scale.

Visual Storytelling, Automated​

Article-led runs get auto-generated cover and inline images rendered via Replicate. Your content looks as good as it reads, straight out of the CLI.

Built for Agents and Automation​

Ideon is designed to slot into modern agentic and CI workflows:

  • MCP server — Expose Ideon tools to Claude Code, ChatGPT, Gemini, or any MCP host
  • Agent runtime registration — ideon agent install registers integration profiles for supported platforms
  • Non-interactive mode — ideon write --no-interactive removes all prompts for CI and automation
  • Machine-readable config — ideon config list --json for agent inspection and orchestration
  • Resumable runs — Pick up exactly where you left off with ideon write resume

Google Keyword Planner Integration​

Query real keyword data from Google Ads directly from the CLI or through MCP tools:

  • Keyword ideas — Generate related keywords from seed keywords, a URL, or a site
  • Historical metrics — Get search volume, competition, and CPC data for any keyword
  • Forecast data — Project impressions, clicks, and cost for keyword campaigns
  • Query history — List cached research runs with filters using ideon gkp list
  • Cache controls — Reuse fresh cached responses by default and force live refresh with --refresh
  • Editorial context — Tag GKP queries with --publication and --series
ideon gkp ideas --keywords seo,marketing --country US
ideon gkp historical --keywords seo --json
ideon gkp forecast --keywords seo --match-type EXACT --country US
ideon gkp list --publication tech-blog --verbose

Set up credentials once with ideon gads login, then query keyword data from the CLI or expose it to any MCP-compatible agent.

Ready to Ship More Content?​

Get Started →

Or jump straight to the CLI Reference and Writing Guide.