Social Media Content Planner

Research-driven monthly social media content calendar planner. Onboards a client, researches trending topics in their niche, runs human approval loops, and delivers a fully scheduled calendar ready for content production.

Spotlight Revenue@spotlight-revenue

Install

openclaw skills install @spotlight-revenue/social-content-planner

Social Media Content Planner — SKILL.md

What This Skill Does

Takes a client from zero to a fully researched, human-approved monthly content calendar. The agent:

Runs a structured onboarding questionnaire to understand the client's niche, audience, content types, and goals
Builds a Google Spreadsheet with dedicated tabs for research, content planning, and the calendar
Researches trending topics in the client's niche using web search + YouTube Data API
Runs a two-gate human approval loop (topic angles → content list) before finalizing the calendar
Produces a ready-to-use monthly calendar the production team can execute against
On the monthly loop, re-runs research and generates a new month without re-onboarding

The spreadsheet is the single source of truth. The human manager approves via conversation — the agent handles all sheet editing.

Prerequisites

Before running this skill, ensure:

Google Sheets access — Google Sheets API v4 via a service account, API key, or Maton connection if using OpenClaw infrastructure
YouTube Data API key — free tier via Google Cloud Console (10,000 units/day). Required for research phase. If unavailable, the agent falls back to web search only.
Communication channel — Slack channel or any messaging surface for the approval loop. For public/standalone use, approval happens in-conversation.

Trigger

Use this skill when asked to:

Plan a month of social media content for a client or brand
Build a content calendar
Research trending topics in a niche for content planning
Run the monthly content planning loop for an existing client
Onboard a new client for social content planning

Phase 0 — Client Onboarding

Run this once per new client. If a Client Config tab already exists in the client's spreadsheet, skip to Phase 1.

Step 1: Run the Questionnaire

Ask the client these questions one at a time (or in a grouped batch if they prefer). Be conversational — not a form dump.

Q1 — Brand basics

"What's the name of the brand or company we're planning content for?"

Q2 — Niche and industry

"In one or two sentences, describe what this brand does and who it serves. What industry or niche is it in?"

Q3 — Target audience

"Who is the ideal person watching or reading this content? Describe them: age range, profession or lifestyle, what problems they have, what they care about."

Q4 — Brand voice and tone

"How should this brand sound? Pick any that apply, or describe it in your own words:

Educational / expert

Casual and relatable

Motivational / high energy

Professional / authoritative

Entertaining / personality-driven

Something else?"

Q5 — Content types

"What kind of content are we planning? Select all that apply:

Short-form video (~60 seconds) — posts to TikTok, Instagram Reels, Facebook Shorts, YouTube Shorts, LinkedIn

Long-form video (5+ minutes) — YouTube

Blog / articles — website and LinkedIn

Image or carousel posts — Instagram, LinkedIn, Facebook

Something else?"

Q6 — Platform confirmation

(Auto-populate based on Q5 answers, then confirm) "Based on your content types, we'll be planning for: [list platforms]. Does that look right, or do you want to add or remove any platforms?"

Q7 — Posting frequency

"How many pieces of content do you want to publish per month? You can give a total, or break it down by content type (e.g., 8 short videos + 2 articles)."

Q8 — Competitors and inspiration

"Name 3–5 accounts, creators, or brands that are doing content you admire or compete with. They don't have to be in your exact niche — just accounts that set the standard for what good looks like."

Q9 — Off-limits

"Are there any topics, angles, or types of content that are off-limits for this brand? Anything we should never touch?"

Q10 — Primary goal

"What's the main thing you want this content to accomplish? Pick one:

Build brand awareness

Generate leads or sales

Educate the audience

Build community and engagement

Establish authority in the niche

Something else?"

Step 2: Create the Spreadsheet

After the questionnaire is complete, create a new Google Spreadsheet named: [Brand Name] — Content Planner

Create the following tabs:

Client Config — populated immediately with onboarding answers
Research [Month YYYY] — blank, ready for Phase 1
Content List [Month YYYY] — blank, ready for Phase 3
Calendar [Month YYYY] — blank, ready for Phase 5

Share the sheet with edit access (anyone with the link). Report the sheet URL to the human manager.

Client Config Tab Structure

Field	Value
Brand Name
Niche / Industry
Target Audience
Brand Voice
Content Types
Platforms
Posts Per Month
Posts Per Month (breakdown)
Competitor / Inspiration Accounts
Off-Limits Topics
Primary Goal
Sheet Created	[date]
Last Updated	[date]

Phase 0.5 — Seed Mode (Optional)

After onboarding (or at the start of any monthly loop), ask before running research:

"Do you have source content you'd like me to pull ideas from — an ebook, transcript, recording, or a list of ideas you already know you want? Or should I go straight to web research?"

If the client provides seed content:

Accepted seed formats:

Ebook or document — paste text or provide a file/URL
YouTube video — provide a URL; extract the transcript via https://www.youtube.com/watch?v=[ID] and summarize key topics
Zoom/Meet recording or transcript — paste the transcript or upload the file; extract every topic, question, or theme raised in the call
Bullet ideas — a short list of topics the client already wants to cover

Seed Processing Steps:

Ingest the seed — accept whatever format the client provides (text paste, file, URL, bullet list)
Extract content angles — pull distinct topic angles from the material. Treat every major point, question, or theme as a potential content angle. Aim for at least 5–10 angles per seed.
Write to Research tab — add seed-derived angles to the Research [Month YYYY] tab with:
- Signal Source: Seed — [type: transcript / ebook / ideas / etc.]
- Status: Pending Approval
Ask the client: "Do you want to mix these with web research results, or use seed-only for this month?"
- Mix: Run Phase 1 normally, then merge seed angles + research angles in the Research tab
- Seed-only: Skip Phase 1 and go straight to Phase 2 with the seed angles

Archiving displaced research angles:

If seed ideas are added to a calendar that already has research-based content planned:

Mark any research angles being replaced with Status: Archived in the Research tab (never deleted)
Insert the seed-derived angles as new rows with Status: Pending Approval
Re-run Phase 2 approval loop for the updated list before updating the calendar

Phase 1 — Research

Run each month. Read Client Config for niche, target audience, content types, and competitor accounts before starting.

Step 1: Web Search Battery

Run all of the following search queries, substituting [niche] with the client's niche and [month year] with the current planning month:

text

[niche] trending topics [month year]
[niche] trending on Twitter [month year]
[niche] what's blowing up on social media right now
[niche] viral content ideas [year]
[niche] top YouTube videos [month year]
[niche] content ideas for [target audience description]
[niche] most shared articles [month year]
[competitor account name] best performing content

Add 2–3 more searches if the niche is specific enough to warrant narrower queries (e.g., a local service business might also search [city] [niche] trending).

Collect all raw signals. Don't filter yet.

Step 2: YouTube Data API Research

If a YouTube Data API key is configured, run the following call:

text

GET https://www.googleapis.com/youtube/v3/search
  ?part=snippet
  &q=[niche keywords]
  &type=video
  &order=viewCount
  &publishedAfter=[30 days ago, ISO 8601]
  &maxResults=25
  &key=[YOUTUBE_DATA_API_KEY]

Extract: video titles, channel names, view counts (approximate from metadata). Identify recurring topic patterns and angles across the top results.

If no API key is configured, note this and proceed with web search results only.

Step 3: Synthesize Into Topic Bank

Combine web search signals + YouTube findings into a clean list of 20–30 topic angles. For each angle, write:

Topic angle — specific framing, not just a broad subject (e.g., "Why most [niche] businesses are leaving money on the table with X" not just "X")
Signal source — what made this show up (trending search, YouTube views, competitor content, etc.)
Content type fit — which content types this angle works best for (short video, article, long-form, etc.)

Write this list to the Research [Month YYYY] tab with columns: #, Topic Angle, Signal Source, Content Type Fit, Status (default: Pending Approval).

Phase 2 — Topic Approval Loop

Step 1: Post Summary to Channel

Post the topic list to the Slack channel (or in-conversation for standalone use) in this format:

text

Here are the [N] topic angles I've researched for [Brand Name] — [Month].

Review the full list in the sheet: [sheet URL]

Here's a quick summary of the top angles:

1. [Topic angle] — [1-line rationale]
2. [Topic angle] — [1-line rationale]
3. [Topic angle] — [1-line rationale]
... (top 10–12 in the message, full list in the sheet)

Let me know:
- Which ones you want to keep, cut, or adjust
- Any topics you'd add that aren't here
- Any angle adjustments needed

Step 2: Revise Based on Feedback

When the human manager responds:

Mark rejected angles as Rejected in the Status column
Update any angles per feedback
Add any new angles the manager suggests
Re-post the updated summary if significant changes were made

Step 3: Confirm Approval

When the human manager explicitly approves (e.g., "looks good," "approved," "let's go with these"), mark all remaining Pending Approval angles as Approved in the sheet and confirm in the channel:

"Topic list approved. Moving on to content list generation."

Phase 3 — Content List Generation

Read the approved topic angles from Research [Month YYYY]. Read the Posts Per Month target and content type breakdown from Client Config.

For each content slot in the month, generate a specific post concept:

Column	Description
#	Slot number
Working Title	Specific, compelling title — not generic
Content Type	Short video / Long video / Article / Image / etc.
Platform(s)	Which platforms this publishes to
Hook / Angle	One sentence — the specific frame that makes this worth watching/reading
Source Topic	Which approved topic angle this comes from
Priority	Suggested publish order (1 = first of month)
Status	`Draft` by default

Fill the Content List [Month YYYY] tab completely. Every slot in the month should have a row.

Phase 4 — Content List Approval Loop

Same mechanic as Phase 2.

Post to channel:

text

Content list for [Brand Name] — [Month] is ready for review.

Full list in the sheet: [sheet URL]

Here's the lineup:

[Priority 1] [Working Title] — [Content Type] — [Hook/Angle]
[Priority 2] [Working Title] — [Content Type] — [Hook/Angle]
... (first 8–10 in the message, full list in the sheet)

Total: [N] pieces across [content types]. Let me know any changes.

Revise the Content List tab based on feedback. Repeat until explicit approval.

On approval, confirm:

"Content list approved. Building the calendar now."

Phase 5 — Calendar Build

Read the approved Content List [Month YYYY] tab. Map each piece of content to a specific publish date.

Rules for calendar distribution:

Spread content evenly across the month — avoid clustering
For multi-platform same-content (e.g., one short video → TikTok + IG + FB + YT + LinkedIn), this counts as one content piece, one row, one date
Vary content types across the week where multiple types are in the mix
Priority 1 = first publish date, not necessarily Day 1 of the month (account for production time — assume at least 3–5 days lead from calendar confirmation to first publish)

Calendar Tab Structure

Date	Day	Working Title	Content Type	Platform(s)	Hook / Angle	Priority	Status
YYYY-MM-DD	Monday					1	Planned

Write the full calendar to the Calendar [Month YYYY] tab.

Post to channel:

text

Calendar is built for [Month]. Here's the overview: [sheet URL]

First publish: [date] — [title]
Total pieces: [N]
Content mix: [X short videos, Y articles, etc.]

Review the dates and let me know if anything needs shifting. Otherwise this is ready for production.

Phase 6 — Handoff to Production

The Google Spreadsheet is the handoff artifact.

When the human manager is ready to begin content production (script writing, article writing, etc.), they start a new session with the relevant production agent and provide:

"Here's our content calendar for [Brand Name]: [sheet URL] Start with Priority 1 and work down. We're producing [content type]."

No additional file needed. The Calendar tab has everything the production agent needs.

Phase 7 — Monthly Loop

When the human manager returns for the next month's planning:

Read the existing spreadsheet URL from the client record
Read Client Config — no re-onboarding needed
Check previous month's Calendar tab to note topics already covered (avoid immediate repeats)
Add three new tabs for the new month: Research [Month YYYY], Content List [Month YYYY], Calendar [Month YYYY]
Run Phases 1–6 for the new month

Each month builds on the previous one. The sheet accumulates a full history of all content planned for the client.

Quick Reference: Spreadsheet Tab Structure

Tab Name	Created When	Purpose
`Client Config`	Onboarding	Stores all client configuration — never deleted
`Research [Month YYYY]`	Start of each month	Trending topic bank, approval status
`Content List [Month YYYY]`	After topic approval	Specific post concepts, titles, hooks
`Calendar [Month YYYY]`	After content list approval	Final dated calendar, production status

Quick Reference: YouTube Data API Setup

Go to Google Cloud Console
Create or select a project
Enable "YouTube Data API v3"
Create credentials → API Key
Add the key as YOUTUBE_DATA_API_KEY in your environment or pass it to the agent at session start

Free quota: 10,000 units/day. A typical research run uses ~100–300 units.

Notes for Public Skill Users

Replace Maton Google Sheets calls with direct Google Sheets API v4 calls using your own credentials
Approval loop defaults to in-conversation — no Slack required
YouTube Data API key is optional but strongly recommended for richer research
The skill is client-agnostic — run it for any brand in any niche