- Home
- Skills
- Social Media
- Social Media Department: One Idea, 8 Platforms, 1 Approval
Works with the AI Agents you already use
Social Media Department: One Idea, 8 Platforms, 1 Approval
Give it one idea. Get native posts for up to 8 platforms, audited before you see them, approved in one pass, and loaded into your scheduler as drafts. Nothing posts without your OK.
$19
Social Media Department: One Idea, 8 Platforms, 1 Approval
Example session with this skill installed
Turn this into posts for Instagram, LinkedIn and X:
We're a two-person candle shop. Our 8 oz soy candle kept tunneling, so we tested four wick sizes at the kitchen table over two weeks. The third one finally melted edge to edge. New batch ships Monday.
- Read your context and instructions
- Compiled the social media department
Repurposing set · Wick test · sample
Source: owner's note · Pillar: behind the scenes · Platforms: Instagram, LinkedIn, X
Atoms
| # | Atom | Type | Needs claim validation? |
|---|---|---|---|
| 1 | Our 8 oz soy candle kept tunneling | problem | No (your own report) |
| 2 | Four wick sizes, two weeks, kitchen table | process | No |
| 3 | Wick 3 melts edge to edge | result | Yes: send your burn-test notes |
| 4 | New batch ships Monday | date | Yes: confirm on post day |
Map
| Platform | Format | Atoms | Angle | One job | Suggested day |
|---|---|---|---|---|---|
| carousel, 5 slides | 1-3 | show the four test candles side by side | save | Tue | |
| text post | 1, 2 | why a two-person shop spent two weeks on one wick | reply | Wed | |
| X | single post | 2-4 | the sharpest line | click | Thu |
Drafts
Instagram · carousel
Hook (slide 1): Four wicks. Two weeks. One winner.
Slides 2-5: one candle each after the same burn time, labeled Wick 1-4; Wick 3 marked "the keeper".
Caption: Our 8 oz soy candle kept tunneling, so we tested four wick sizes at the kitchen table. Swipe to see which one melted edge to edge. Save this if you've ever lost half a candle to a tunnel.
Alt text (slide 1): Four amber jar candles in a row on a wooden table, each with a different wick.
Hashtags: a few niche tags, within the current limit in the Instagram playbook.
Claim refs: atom 3
LinkedIn · text post
Hook: We spent two weeks on one wick.
Post: Two of us run this candle shop. Our 8 oz soy candle kept tunneling, so before the next batch we tested four wick sizes, one at a time. Wick 3 melted edge to edge. Slow for a shop our size, but it's why we can stand behind the new batch. What's one thing you test before you ship?
Claim refs: atom 3
X · single post
Post: Four wicks, two weeks, one kitchen table. Wick 3 finally melted our 8 oz soy candle edge to edge. New batch ships Monday. [link]
Link: product page, with UTM tags if you use them
Claim refs: atoms 3, 4
Assets to supply
- 5 photos: all four candles together, then each one after the same burn time, same light
- Your burn-test notes (atom 3), and the ship date confirmed on post day (atom 4)
Next: audit by the launch-gatekeeper in a separate chat, then into this week's approval file. Nothing is scheduled until you approve it.
Filed: 11_channels/social/drafts/2026-11-02_wick-test.md
Connects securely to your tools. The creator never sees your data.
About this skill
Posting everywhere shouldn't mean writing everything. Give it one idea and get back a week of posts, each written for the platform it's going on, checked, and waiting for one yes.
What changes
- Before: One caption pasted on every platform. After: A native version for each platform you use (up to 8), written for how people use it.
- Before: Guessing at caption limits and video specs. After: 8 playbooks with current specs, each checked in October 2026 with the official source linked.
- Before: Copy-pasting posts into your scheduler. After: Approved posts land in Buffer, Typefully, Postiz or Eden as drafts, or you get an import file.
- Before: Posting something that overpromises. After: A gatekeeper audits every draft before you see it.
Why it works
Built by Tom Sun, who grew a consumer brand's revenue from about $3M to $6M as its senior director of marketing, and co-founded PAVE Agency (120+ brands). It's spun off the One-Person Marketing Department, built from the AI team he runs for that brand, so it follows the same rule: the team proposes, you publish.
Everything you get
- Weekly system. Monday plan, mid-week drafts, one approval file, Friday recap from your own analytics.
- Content repurposer. One idea becomes a native post for each platform: Instagram, TikTok, X, LinkedIn, Threads, YouTube Shorts, Facebook and Pinterest.
- 8 platform playbooks. Caption limits, hashtag rules, video lengths and aspect ratios, each with its source and the date it was checked.
- 8 scheduler guides. Connection steps for Buffer, Typefully, Postiz, Eden, Metricool, Hootsuite, Ayrshare and Later, plus an export file and phone checklist if you don't use one.
- Launch Gatekeeper. Every draft is checked for unproven claims, fake urgency and missing disclosures.
- 5 templates. Brand brain, claim ledger, decisions log, content calendar and hook bank.
How fast
Say "Set up my social media." Tell it one thing you want to post this week, and you get audited drafts for your platforms in the same session.
The price
$19 once. No subscription. Want the full 30-role team too? The bundle with the Marketing Department saves 20%.
Launch bonus
First 10 buyers get a free 30-minute setup call with me. After you buy, send me a message here on Agensi to book it.
Guarantee
Covered by Agensi's money-back guarantee: if it doesn't do what this page says, you get your money back.
For one-person businesses that would rather approve posts than write them. Not for auto-posting bots, follower services or engagement pods.
Works with Claude.ai, the Claude desktop app, Cowork and Claude Code. A scheduler is optional.
Built not to post on its own, use bots or follower services, or invent statistics or "best time to post" numbers.
It never posts, boosts or replies publicly on its own; scheduler items are drafts or held posts. Specs were checked in October 2026 and platforms change them, so each file is dated with its sources; check anything critical in the app. No 'best time to post' numbers: recaps use your own analytics. Later has no Claude connector, so the pack exports a file for it. On Claude.ai and desktop, Skills need a paid plan with code execution turned on.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 18 days ago
Requires
Permissions
Read/Write Files: the marketing/ folder (brand brain, claim ledger, calendar, drafts, export files). Network Access: only if you connect a scheduler or analytics tool. Scheduler items are drafts or held posts; nothing posts without your OK.