naluforge/ Founder OS
Content Ideation Engine | Founder OS

Ideate

Content ideation engine that generates drafts, outlines, variations, and research briefs in authentic founder voice.

What Does the Content Ideation Engine Do?

The Ideate namespace is your content creation starting point. It takes a topic, document, or research question and produces structured content briefs, full drafts, and multiple angle variations -- all written in an opinionated founder voice calibrated for LinkedIn, X, Meta, and TikTok. The output considers platform norms but does not apply platform-specific formatting; it focuses on the creative substance that makes content worth reading.

Six commands cover the full ideation pipeline. Start with research to discover data-backed angles and supporting statistics. Use outline to generate a structural scaffold with hook options and framework selection. Run draft to produce a complete content piece in one pass. Feed an existing document into from-doc to extract key points and generate social-ready angles. Use variations to explore multiple approaches to the same topic with different hooks, frameworks, and tones. And use facts to pull quotable statistics and data points from any document or URL.

Every command applies a consistent founder voice: professional but conversational, opinionated with evidence, short-long sentence rhythm, and concrete examples over abstract advice. Seven content frameworks (story, listicle, contrarian, how-to, lesson, insight, question) provide structural variety, and four audience segments (founder, technical, marketer, CXO) calibrate tone and examples.

Required Tools

Tool Required Purpose
None -- Most ideate commands run without external tools
Web Search MCP Optional Powers the research command for live data and trend analysis
WebFetch Optional Powers the facts command for URL-based fact extraction
Filesystem Optional Save drafts to file via --to-file flag

Commands

/founder-os:ideate:outline

What it does -- Generates a structured content outline from a topic, selecting the best framework and adapting structure to the target platform. Produces hook options, body section scaffolding, and closer variants. Available in skeleton mode (quick scaffold) or detailed mode (beat-by-beat with transition notes and word counts).

Usage:

/founder-os:ideate:outline "why founders should learn to code"
/founder-os:ideate:outline "hiring your first engineer" --framework=howto --platform=linkedin --depth=detailed
/founder-os:ideate:outline "AI replacing jobs myth" --framework=contrarian --platform=x

Example scenario:

You have a rough idea about why you stopped using OKRs but are not sure how to structure it. You run /founder-os:ideate:outline "why I stopped using OKRs". The system auto-selects the "personal lesson" framework, generates 2 hook options (a story-entry hook and a contrarian hook), maps out 4 body sections, and offers 2 closer variants. You pick the contrarian hook and proceed to drafting.

What you get back:

A structured outline with topic, selected framework (with reasoning if auto-selected), platform, and depth. Includes 2-3 hook options labeled by formula type, body sections adapted to platform norms (3-5 for LinkedIn, 3-7 thread beats for X), and 2-3 closer options (CTA, reflection, callback). In detailed mode, each section includes bullet points, transition notes, and estimated word counts.

Flags:

  • --framework=FRAMEWORK -- Content framework: story, listicle, contrarian, howto, lesson, insight, question, or auto (default: auto)
  • --platform=PLATFORM -- Target platform: linkedin, x, meta, tiktok (default: linkedin)
  • --depth=DEPTH -- Outline depth: skeleton (default) or detailed

/founder-os:ideate:draft

What it does -- Generates a complete content draft in one pass through a three-phase pipeline: ideation (framework selection, hook generation, audience calibration), drafting (full content in founder voice), and validation (quality checklist, character count). Output is a platform-agnostic content brief ready for formatting via /social:compose.

Usage:

/founder-os:ideate:draft "why I stopped using OKRs"
/founder-os:ideate:draft "hiring your first engineer" --audience=founder --framework=story
/founder-os:ideate:draft "AI tools for small teams" --platform=x --length=long
/founder-os:ideate:draft "cold email tips" --framework=listicle --to-file=drafts/cold-email.md

Example scenario:

You need a LinkedIn post about a lesson you learned when your best engineer quit. You run /founder-os:ideate:draft "what I learned when my best engineer quit" --framework=lesson --audience=founder. Phase 1 selects a story-entry hook, identifies 4 key points, and calibrates for founder audience. Phase 2 writes the full draft in founder voice with short-long rhythm. Phase 3 validates quality and counts characters. You get a complete content brief with 3 hook candidates and the full draft ready to post.

What you get back:

A content brief with metadata (topic, framework, audience, platform, character count), 3 hook candidates with the selected one marked, and the full content body written in founder voice. If --to-file is specified, the brief is saved with YAML frontmatter.

Flags:

  • --platform=PLATFORM -- Target platform hint (default: linkedin)
  • --audience=SEGMENT -- Target reader: founder, technical, marketer, cxo (default: founder)
  • --framework=FRAMEWORK -- Content framework (default: auto)
  • --length=MODE -- Content length: short, medium, long (default: medium)
  • --tone=TONE -- Humanization tone: professional, friendly, casual, authoritative, conversational (default: professional)
  • --to-file=PATH -- Save the draft to a file

/founder-os:ideate:variations

What it does -- Generates multiple content angle variations for a topic or existing draft, each varying the hook style, framework, and tone. Useful for exploring different approaches before committing to a direction. Ephemeral output only -- no files saved.

Usage:

/founder-os:ideate:variations "why I stopped using OKRs"
/founder-os:ideate:variations --count=5 --platform=x
/founder-os:ideate:variations --audience=technical

Example scenario:

You drafted a post about remote work but it feels flat. You run /founder-os:ideate:variations (it picks up the draft from the conversation). The system generates 3 variations: a contrarian take with a bold-claim hook, a story framework with a micro-story hook, and a listicle with a stat-led hook. Each variation includes a 2-3 sentence opening paragraph, 3-5 talking points, and a platform fit rating. You pick elements from variations 1 and 3 to combine.

What you get back:

Each variation includes: framework and hook formula labels, a one-sentence angle description, a 2-3 sentence opening paragraph demonstrating the hook and tone, 3-5 key talking points, a suggested structure, and a platform fit note. A comparison table at the end summarizes all variations side by side with fit ratings (Strong/Good/Fair).

Flags:

  • --platform=PLATFORM -- Target platform (default: linkedin)
  • --audience=SEGMENT -- Target audience (default: founder)
  • --count=N -- Number of variations, 1-5 (default: 3)
  • --tone=TONE -- Humanization tone preset (default: professional)

/founder-os:ideate:from-doc

What it does -- Transforms a document, article, or pasted text into content angles and briefs for social media. Extracts key points, auto-classifies the document type, selects the best framework, and generates platform-aware angle recommendations in founder voice.

Usage:

/founder-os:ideate:from-doc blog-post.md
/founder-os:ideate:from-doc ~/Documents/quarterly-report.pdf --framework=insight --audience=cxo
/founder-os:ideate:from-doc meeting-notes.txt --platforms=linkedin,x

Example scenario:

You published a detailed blog post about your hiring process and want to repurpose it for social media. You run /founder-os:ideate:from-doc blog-post.md --platforms=linkedin,x. The system extracts 6 key points, identifies the 3 strongest angles (a contrarian take on job descriptions, a how-to on the interview loop, a story about the worst hire), and generates per-platform recommendations. LinkedIn gets the detailed how-to; X gets the contrarian thread.

What you get back:

A content brief with source metadata, extracted key points (each with a supporting detail and content angle), the core narrative angle, a hook concept, and per-platform recommendations when multiple platforms are specified. If --to-file is specified, the brief is saved with frontmatter.

Flags:

  • --platforms=PLATFORMS -- Comma-separated target platforms (default: linkedin)
  • --audience=SEGMENT -- Target audience (default: founder)
  • --framework=FRAMEWORK -- Content framework (default: auto)
  • --tone=TONE -- Humanization tone preset (default: professional)
  • --to-file=PATH -- Save the content brief to a file

/founder-os:ideate:research

What it does -- Researches a topic to discover content angles, supporting data, and platform-specific hooks. Breaks the topic into searchable dimensions (industry, contrarian, data, trend, personal experience), runs web searches, and synthesizes 3-5 ranked content angles with hook ideas, supporting statistics, and framework recommendations.

Usage:

/founder-os:ideate:research "freelancing vs employment"
/founder-os:ideate:research "AI tools for small teams" --platforms=linkedin,x --depth=deep
/founder-os:ideate:research "remote work trends 2026" --depth=quick

Example scenario:

You want to write about AI adoption in small businesses but need data to back your claims. You run /founder-os:ideate:research "AI adoption small business" --depth=deep. The system runs 10 targeted searches, finds recent statistics on adoption rates, discovers 3 competitor posts on the topic, and synthesizes 5 content angles. Angle #1: a stat-led post using a counterintuitive finding about small teams outperforming enterprises on AI implementation.

What you get back:

A research brief with 3-5 ranked content angles. Each angle includes a hook idea, 2-3 supporting data points with sources, a platform fit rating per requested platform, a recommended content framework, and a one-sentence explanation of why the angle works. If WebSearch is unavailable, angles are based on general knowledge with a degraded-mode notice.

Flags:

  • --platforms=PLATFORMS -- Comma-separated target platforms (default: linkedin)
  • --depth=DEPTH -- Research depth: quick (3-5 searches) or deep (8-12 searches with competitor analysis) (default: quick)

/founder-os:ideate:facts

What it does -- Extracts facts, statistics, and quotable data from a document or URL, classifies each by its content role (hook-worthy, body-supporting, credibility-building), and presents them ranked by impact. Useful for backing your content with credible evidence.

Usage:

/founder-os:ideate:facts quarterly-report.pdf
/founder-os:ideate:facts research-paper.md --format=table --max=15
/founder-os:ideate:facts https://example.com/industry-report

Example scenario:

You found an industry report with useful data for your next LinkedIn post. You run /founder-os:ideate:facts industry-report.pdf --max=10. The system extracts 14 facts, classifies them, and returns the top 10: 3 hook-worthy statistics (surprising enough to open a post), 4 body-supporting data points (evidence for claims), and 3 credibility-building citations (named studies and expert quotes).

What you get back:

Facts organized by classification: hook-worthy facts (surprising stats for opening a post), body-supporting facts (evidence for building an argument), and credibility-building facts (authoritative sources and expert quotes). Each fact includes the data point and its source attribution. Available in bullet or table format.

Flags:

  • --format=FORMAT -- Output format: bullets (default) or table
  • --max=N -- Maximum number of facts to return (default: 10)

What Content Frameworks Does Founder OS Support?

Seven content frameworks provide structural variety: Story (personal narrative with a lesson), Listicle (numbered takeaways), Contrarian (challenge a popular belief), How-To (step-by-step guide), Lesson (what you learned from experience), Insight (data-backed observation), and Question (pose a provocative question and answer it). When set to "auto," the system selects the best framework based on your topic, audience, and platform.

Tips & Patterns

  • Research-to-publish pipeline: Run ideate:research to find angles, then ideate:draft to write, then /social:compose to format for platform, then /social:post to publish. Each step builds on the previous.
  • Repurpose everything: Wrote a blog post? Run ideate:from-doc. Had a great meeting? Extract angles from the notes. Found a report? Run ideate:facts to pull quotable data.
  • Explore before committing: Use ideate:variations after a draft to see if there is a stronger angle you missed. The comparison table makes it easy to pick the best elements.
  • Match framework to platform: Contrarian takes and question-led posts perform well on X. Story and how-to frameworks shine on LinkedIn. Let auto-selection guide you, or override when you know your audience.
  • Build a content bank: Use --to-file on draft and from-doc commands to save content briefs locally. Review and schedule them over the coming week.
  • Social -- The publishing layer; social:compose formats ideate output for specific platforms, social:post publishes it
  • Knowledge Base -- KB documents can be fed into ideate:from-doc to generate social content from internal knowledge
  • Newsletter -- Newsletter research and drafting complements ideation; newsletter content can be repurposed via ideate:from-doc