source-command-content-ideation-linkedin

name	source-command-content-ideation-linkedin
description	Generate LinkedIn posts from Content Engine + daily notes + web research

source-command-content-ideation-linkedin

Use this skill when the user asks to run the migrated source command content-ideation-linkedin.

Command Template

LinkedIn Post Generation

Generate high-performing LinkedIn posts.

Config

Setting	Value
Default Count	5
Output	Dynamous/Content-Ideation/LinkedIn/YYYY-MM-DD/

Paths (from dynamous-engine root)

Input:

content-engine/output/DAILY_DIGEST.md        ← Read first
content-engine/output/topics/                ← List folders, read most recent
Dynamous/Daily Notes/                        ← List files, read most recent (has a space!)

Skill Reference (read these for templates/guidelines):

.Codex/skills/linkedin-post/SKILL.md
.Codex/skills/linkedin-post/references/style-guide.md

Output:

Dynamous/Content-Ideation/LinkedIn/YYYY-MM-DD/

Workflow

1. Read Skill Guidelines

Read these files to understand LinkedIn post philosophy and voice:

• .Codex/skills/linkedin-post/SKILL.md
• .Codex/skills/linkedin-post/references/style-guide.md

These define: content modes (story, observation, take, teach, react, ask), specificity requirements, anti-patterns to avoid, and voice guidelines.

2. Read Content Engine Data

Read content-engine/output/DAILY_DIGEST.md

List content-engine/output/topics/ folders, sort by name, read all .md files from the most recent date folder.

3. Read Daily Notes

List Dynamous/Daily Notes/ files (format: YYYY-MM-DD.md), sort by name, read the most recent one.

Look for IMPORTANT markers (must include) vs general ideas (context only).

4. Web Research

Search 5-7 key topics from the digest for current news and trends.

5. Prioritize

1. IMPORTANT items from daily notes (must include)
2. Digest items (primary source)
3. General daily note ideas (use to weight digest items)

6. Generate Posts

Get today's date. Create folder: Dynamous/Content-Ideation/LinkedIn/YYYY-MM-DD/

For each selected idea, create a post following the linkedin-post SKILL.md guidelines:

• Include Source Attribution at the top (digest item #, original URL from topic file, deep dive path)
• Choose appropriate content mode (story, observation, take, teach, react, or ask)
• Include at least 2 concrete details (dates, numbers, names, specific context)
• Write in authentic voice - no template phrases or engagement bait
• No hashtags
• End naturally - no formulaic CTAs

Write each post to Dynamous/Content-Ideation/LinkedIn/YYYY-MM-DD/{slugified-topic}.md

7. Create Index

Create Dynamous/Content-Ideation/LinkedIn/YYYY-MM-DD/index.md:

# LinkedIn Posts - YYYY-MM-DD

## Posts Created

### 1. {Topic}

- **File**: [{slug}.md](./{slug}.md)
- **Mode**: {story/observation/take/teach/react/ask}
- **Core insight**: {one sentence summary}

## Ideas for Later

- {remaining ideas not used this batch}

Output Structure

Dynamous/Content-Ideation/LinkedIn/2026-01-03/
├── index.md
├── ai-agent-orchestration.md
├── developer-divide-opinion.md
├── rag-optimization-tips.md
├── Codex-opus-review.md
└── memory-scaling-insight.md

Platform Notes

• Extract single insight per post, don't summarize entire topics
• First 2 lines matter, but don't use formulaic hooks - be specific and authentic
• Links are fine in post body (one max)
• No hashtags
• No engagement bait CTAs ("Agree?", "What would you add?")
• Posts should sound like you talking to a colleague, not like "a LinkedIn post"