04_linkedin_content

This is the operating system for Stacksync's LinkedIn presence. Every post we write, publish, analyze, or archive lives here. One folder per persona, consistent naming everywhere.

Daily Workflow — Start Here

When posts go live on LinkedIn

1. Publish each post Tell Claude: "this post was published: [persona] [linkedin-url]" Or use the skill: /publish-post [persona] [linkedin-url]

Claude will:

• Move the draft .md + image from drafts_[Persona]/ → published_[Persona]/
• Update the YAML frontmatter (status, published_date, link, image)
• Rename the image to match the post slug
• Delete the original draft
• Push to git

2. Generate comments Tell Claude: "do the comments for all the posts"

Claude reads each published post and generates 3 comment options per commenter. Saved to: 00_comments/YYYY-MM-DD_comments.md

The 5 personas comment on each other's posts. The post author never comments on their own post.

3. Run the commenting session Tell Claude: "run it"

Claude opens post #1 in all commenter Brave profiles simultaneously. You comment on all tabs → tell Claude "next" → Claude opens post #2. Repeat.

You say	Claude does
"run it"	Opens post #1 in all profiles
"next"	Opens the next post
"open post 3"	Jumps to a specific post
"run it for 2026-02-25"	Uses a specific date

Script location: scripts/open_comments.py Full process doc: 00_comments/COMMENTING_PROCESS.md

---

Brave Profile Map

Persona	Brave Profile
Ruben	Profile 5
Nacho	Profile 6
Alexis	Profile 4
Carter	Profile 7
Brittany	Profile 10

---

Folder Map

00_comments/          ← Daily comment files (YYYY-MM-DD_comments.md)
scripts/              ← open_comments.py (Brave automation)
skills/               ← post-commentator.skill (comment generator)
[Persona]/
  drafts_[Persona]/   ← Active drafts
  published_[Persona]/← Live posts + images/
  [name]_personality_v01.md

---

How It Works

Each persona on the team has their own folder. Inside, everything follows the naming pattern type_PersonaName:

04_linkedin_content/
│
├── Ruben/                  CEO — owns founder stories, industry deep dives
├── Alexis/                 CTO — owns technical content, data engineering
├── Nacho/                  GTM — owns go-to-market, sales ops, SDR content
├── Tony/                   Industry Veteran — owns enterprise ops perspective
├── Carter/                 Young Commercial — owns early-career B2B angle
│
└── accounts_research/      External creator analysis (16 creators, Kallaway method)

Inside Each Persona Folder

Folder	What Goes Here	Naming
published_[Name]/	Every original post that went live on LinkedIn	YYYY-MM-DD_slug.md
bangers_[Name]/	Published posts with 50+ total engagement	BANGER_[NAME_]YYYY-MM-DD_slug.md
drafts_[Name]/	Active drafts being worked on right now	descriptive_name_persona.md
research_[Name]/	LinkedIn export CSVs, analysis scripts, Kallaway insights	Datasets + scripts
[name]_personality_v01.md	The persona's voice, style, hooks, and content lens	At folder root

Post File Format (YAML Template)

Every published post follows this template (00_ai_markdowns/17_simple-linkedin-post-template.md):

---
name: YYYY-MM-DD_slug_of_first_line
description: "Author published post - First line of post."
domain: business
node_type: linkedin-post
status: published
last_updated: YYYY-MM-DD
published_date: YYYY-MM-DD
tags:
  - integration
  - data-sync
topics:
  - "Integration/Data"
  - "Teacher Hook"
related_concepts:
  - "[[persona_personality_v01]]"
link: https://www.linkedin.com/posts/...
image: images/YYYY-MM-DD_slug.jpg
---

<!-- Kallaway Analysis -->
<!-- Rank: #X | Engagement: N (Likes: N, Comments: N, Shares: N) -->
<!-- Hook Type: Teacher | Topic: Integration/Data | Angle: Data/Numbers | Structure: Story → Lesson -->
<!-- Original [Persona] post: True -->

Post text goes here.

Status values: draft | review | published

Bangers — 50+ Engagement Rule

A post enters bangers_[Name]/ only when:

1. It is an original post by the persona (not a reshare)
2. It has 50+ total engagement (likes + comments + shares)
3. The MD is copied from published_[Name]/ with a BANGER_ prefix
4. The image is copied to bangers_[Name]/images/

Bangers are populated from the Kallaway CSV inside published_[Name]/ — that CSV ranks every post by engagement and flags originals.

Kallaway 5 Content Legos

Every post is classified using the Kallaway method (5 dimensions):

Lego	Options
Hook Type	Teacher, Contrarian, Magician, Fortune Teller, Experimenter
Topic	Integration/Data, Industry/SaaS, Startup/Founder, GTM/Sales, Technical/Engineering
Angle	Data/Numbers, Contrarian/Challenge, Personal Experience, Authority/Credibility
Structure	Story > Lesson, Listicle/Breakdown, Before/After, Question > Answer
Visual Format	Text-only, Carousel, Image+Text, Video

Full framework: /Users/TERABYTE10/Downloads/Moi/Kallaway_Clone.md

Processing a New LinkedIn Dataset

When someone exports a new dataset CSV from LinkedIn:

1. Drop the CSV into research_[Name]/
2. Run Kallaway classification on all posts
3. Separate originals from reshares
4. Generate MDs for every original post in published_[Name]/ using the YAML template
5. Download images to published_[Name]/images/
6. Generate [name]_all_published_kallaway.csv ranking all posts by engagement
7. Copy any post with 50+ engagement to bangers_[Name]/

Content Rules

• All content must serve the ICP: 00_ai_markdowns/00_icp_stacksync.md
• Master process: 00_ai_markdowns/00_content_creation_process.md
• "Here's" constructions are banned across all personas
• Em dashes (—) are banned across all personas
• Each persona writes from their own personality doc — never cross-pollinate voices

Current State (Feb 2026)

Persona	Published	Bangers	Active Drafts	Dataset
Ruben	51	10	1	Processed
Alexis	49	6	0	Processed
Nacho	46	3	1	Processed
Tony	—	—	31	Ready
Carter	—	—	0	Ready