agentleFS
lumizonelumizone/social-content-os-claude-skill/skills/social-content-os/SKILL.md

social-content-os

Skill0 starsChanged 5 months agoMITView on GitHub

Router and onboarding for the Social Content OS pack — a system for planning and producing high-performing social media content (Instagram, LinkedIn, TikTok, X) based on the Interest Graph / Graffiti playbook (ABC hooks, SGTMS, Awareness Funnel, Best of Nine, carousels, ManyChat monetization). Use when the user wants help with social media content but hasn't picked a specific task, or says "social content os", "pomoz mi z social media", "od czego zaczac z contentem", "help me with my social media", "what should I post". Explains the system, loads the brand profile, and routes to the right sub-skill: content-plan, hook-lab, carousel-builder, caption-writer, or repurpose.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Load the brand: Read brand.json in this skill's directory.

    line 40
  2. Distinctive:

    If the request is broad ("help me grow on Instagram"), start with content-plan.

    line 44
  3. Very specific:

    Ask for these, then write brand.json (keep it short, one pass):

    line 50

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Social Content OS

The entry point and router for the pack. Use it to onboard a user, load their brand, and send them to the right tool. Default output language: English (switch to the user's language if their topic or audience is in another language).

The system in one screen

Reach is decided by the Interest Graph, not your follower count: every post is tested on a small audience first and only scales if it performs. So you ship like a graffiti artist — speed and volume over polish — and let data pick winners. Full reference: reference/frameworks.md (read it before doing strategy work).

The five tools

You want to...Use
Plan what to post (calendar, topic tree, funnel mapping)content-plan
Turn one idea into 10+ ranked scroll-stopping hookshook-lab
Build a 7–10 slide carousel and render the slidescarousel-builder
Write a caption, IG/LinkedIn post, or CTAcaption-writer
Break one long-form asset into a batch of postsrepurpose

How to route

  1. Load the brand. Read brand.json in this skill's directory. If it's still the template (brand_name is "Your Brand"), run onboarding (below) first — good output needs a real brand profile.
  2. Match intent to the table above and invoke that skill via the Skill tool. If the request is broad ("help me grow on Instagram"), start with content-plan.
  3. Chain when useful. A typical flow: content-plan → hook-lab on the best ideas → carousel-builder / caption-writer to produce → repurpose to multiply.

Onboarding (first run / template brand)

Ask for these, then write brand.json (keep it short, one pass):

  • Brand name + handle, and the one-liner: what you help people do.
  • Audience: who you're stopping mid-scroll.
  • Topic Tree: your Niche-of-One angle + 3 broad interests + a few subtopics.
  • Voice: 3–5 adjectives.
  • Colors: background, accent, ink (hex). Defaults are fine if unsure.
  • CTA: the comment-trigger WORD and what ManyChat promises in the DM.
  • Platforms in play.

Once saved, confirm and route to the tool that matches what they came for.

Non-negotiable quality bar

Hook first, always. Clear over clever. One idea, one CTA per asset. Specific beats generic. Every asset maps to a funnel stage. No AI tells in human-facing copy (no em-dashes, en-dashes, smart quotes, or "unlock/elevate/dive in/game-changer").

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.