Social Card Gen logo

Social Card Gen

Community
BrianRWagner
social-card-gen

Generate platform-specific social post variants (Twitter/X, LinkedIn, Reddit) from one source input. Works with or without Node.js script. Includes platform reasoning, quality review, and guardrails against cross-posting spam.

Overview

PublisherBrianRWagner
Repositoryai-marketing-claude-code-skills
Skill namesocial-card-gen
Stars
449
Forks
103
Bundled files
8
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 8 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by BrianRWagner on GitHub. Read the source before you install it.

Installation

Install the Social Card Gen AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/BrianRWagner/ai-marketing-claude-code-skills/tree/main/social-card-gen
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/BrianRWagner/ai-marketing-claude-code-skills.git /tmp/ai-marketing-claude-code-skills
mkdir -p .claude/skills
cp -r /tmp/ai-marketing-claude-code-skills/social-card-gen .claude/skills/social-card-gen
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Social Card Gen in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Social Card Gen on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Social Card Gen is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Social Card Generator

Transform one source message into platform-ready social copy for Twitter/X, LinkedIn, and Reddit.

This skill works two ways:

  • With Node.js: Use generate.js for deterministic, automated output
  • Without Node.js: Manual generation path built directly into this skill (no dependencies required)

Mode

Detect from context or ask: "One platform, all platforms, or all platforms with variants?"

ModeWhat you getBest for
quick1 platform (user picks), optimized variantTesting a single channel
standardAll 3 platforms: Twitter, LinkedIn, RedditStandard cross-platform post
deepAll platforms + 3 variants each + A/B test guidanceCampaign testing, maximizing reach

Default: standard — use quick if they specify a platform. Use deep if they're running a content campaign and want options.


⚠️ MANDATORY: 4-Question Intake Before Generating

Ask these before writing a single word of copy:

  1. Source message: What's the core idea, update, or story? (Paste raw text or describe it.)
  2. Tone goal: Informative? Provocative? Humble brag? Conversational? Educational?
  3. Audience: Who needs to see this? (Founders, marketers, developers, general public?)
  4. CTA: What do you want readers to do? (Visit link, reply, follow, share, start a conversation?)

Without these answers, you'll get generic output that won't land on any platform.


Platform Reasoning Rules (Built In — No External Script Needed)

🐦 Twitter/X

  • Character limit: 280 characters (single tweet) or threaded for long-form
  • Tone: Punchy, direct, opinionated. Twitter rewards brevity and conviction.
  • Hook structure: Lead with the most surprising or specific claim. No warm-up.
  • Hashtags: Max 1-2 directly relevant tags. Never stuff.
  • CTA style: Short and embedded. "What's your take?" or "Link in reply."
  • What kills Twitter posts: Long intros, passive voice, corporate speak, hashtag walls.
  • Example hook pattern: "[Counterintuitive claim]. Here's why: [1-2 sentence reason]. [Engaging question]."

💼 LinkedIn

  • Character limit: 3,000 characters for posts. First 2 lines before "See more" are critical.
  • Tone: Professional but personal. Thought leadership with a human angle.
  • Hook structure: First line must create curiosity or state a bold truth — not "excited to share."
  • Hashtags: 3-5 max, at the end. Never in the body copy.
  • CTA style: Invite discussion. "What's your experience with this?" works well.
  • Opener rule: NEVER start with self-promotional language. Lead with a lesson, story, or observation.
  • What kills LinkedIn posts: Cringe openers ("I'm humbled to announce..."), excessive hashtags, no line breaks.
  • Format tip: Short paragraphs (1-2 lines). White space is your friend.

🔵 Reddit

  • Character limit: No hard limit, but long posts need clear structure with headers.
  • Tone: Authentic, curious, community-first. Reddit readers have a finely tuned BS detector.
  • Hook structure: Situational — frame as a question, lesson learned, or genuine experience.
  • Hashtags: Never. Not a thing on Reddit.
  • CTA style: End with an open question that invites the community to share their take.
  • Self-promotion rule: Never lead with your own promotion. Frame everything as a contribution to the community.
  • What kills Reddit posts: Obvious marketing, lack of genuine insight, "check out my [thing]" openers.

Manual Generation Path (No Node.js Required)

Follow this process exactly:

Step 1: Distill the Core Message

From the source input, extract:

  • The ONE key insight or fact
  • The "so what" — why does this matter to the audience?
  • The evidence or story behind it

Step 2: Generate 3 Platform Variants

Twitter/X variant:

[Bold claim or surprising fact — max 1 sentence]

[1-2 sentences of supporting context]

[Engaging question or CTA]

[1-2 relevant hashtags max]

Check: Under 280 chars for single tweet? If not, decide: cut or thread.

LinkedIn variant:

[First line hook — create curiosity, not announcement]

[1-2 line context paragraph]

[Core insight or framework — 2-4 short paragraphs]

[Personal angle or reflection]

[Discussion CTA — question for the audience]

[3-5 hashtags at the end only]

Check: Does the first line work before "See more"? Does it NOT start with "I'm excited to..."?

Reddit variant:

[Title: specific, searchable, not clickbaity]

[Opening: frame as experience, question, or lesson — community-first]

[Body: genuine story or insight with specifics]

[Closing question to invite community input]

Check: Would this feel at home in the target subreddit? No hashtags?


With Node.js (Automated Path)

bash
npm install

# text input
node generate.js --text "We reduced onboarding time by 35% with a checklist." --stdout

# file input
node generate.js --file examples/input-example.md --outdir examples

# URL input (when network is available)
node generate.js --url https://example.com/post --platforms twitter,linkedin --stdout

If generate.js is unavailable, use the Manual Generation Path above — same quality, just manual.


Quality Self-Review Pass (REQUIRED After Generating)

After generating all 3 variants, run this review before delivering:

CheckTwitterLinkedInReddit
Hook strong? (Would you stop scrolling?)✅/❌✅/❌✅/❌
CTA present and clear?✅/❌✅/❌✅/❌
Platform fit? (Tone matches platform norms)✅/❌✅/❌✅/❌
Guardrails pass? (See below)✅/❌✅/❌✅/❌

Flag the weakest variant and explain why: "The Reddit variant is the weakest because it still reads like a LinkedIn post — too polished, not community-first."


Guardrails (Never Violate)

❌ No identical copy cross-posted — every platform variant must be meaningfully different in tone and structure ❌ No hashtag spam on LinkedIn — max 5, always at the end, never in the body ❌ No self-promotional openers on Reddit — community value first, always ❌ No vague hooks — "I have some thoughts on X" is not a hook ❌ No copy that assumes platform context — Reddit readers don't know your LinkedIn audience


Iteration Loop

After delivering the 3 variants:

  1. Ask: "Which variant feels most off? What's not landing?"
  2. If hook is weak → rewrite the first line using a different angle (counterintuitive, numerical, story-led)
  3. If CTA is weak → try a question instead of a directive
  4. If platform fit is off → re-read the platform rules and rewrite from scratch for that platform

Output Format

Deliver:

## Twitter/X
[Final copy — ready to paste]
Character count: [X]/280

## LinkedIn
[Final copy — ready to paste]

## Reddit
Title: [Post title]
[Body — ready to paste]

## Quality Review
- Strongest variant: [Platform] — because [reason]
- Weakest variant: [Platform] — [what to improve if needed]
- All guardrails passed: Yes / No — [note any issues]

Skill by Brian Wagner | AI Marketing Architect | brianrwagner.com

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Social Card Gen AI skill do?

Generate platform-specific social post variants (Twitter/X, LinkedIn, Reddit) from one source input. Works with or without Node.js script. Includes platform reasoning, quality review, and guardrails against cross-posting spam.

Why use Social Card Gen on TypingMind?

Because you install it once and use it with any model. Social Card Gen is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Social Card Gen in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/BrianRWagner/ai-marketing-claude-code-skills/tree/main/social-card-gen. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Social Card Gen?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Social Card Gen?

As many as you like. As long as a model supports skills, you can use Social Card Gen with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Social Card Gen AI skill free?

It is published on GitHub by BrianRWagner. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇