Social Media Research Skills is a set of Agent Skills that turns an AI coding agent into a social media research assistant, aimed at marketers, growth teams, and analysts who want structured research artifacts out of public posts, videos, comments, and ad libraries.

What it is

The project is a collection of thirteen documented skills that an AI agent loads and runs, covering outlier post discovery, transcript analysis, comment mining, competitor teardowns, ad library analysis, trend discovery, influencer prospecting, audience research, social listening, product demand research, creator profile teardowns, content repurposing, and raw endpoint routing. It lives in the agent skills ecosystem, follows the Agent Skills spec published at agentskills.io, and is written in Python under the MIT licence. Each skill pairs a description of the workflow with the output it should produce — an outlier table, a voice-of-customer report, a competitor brief, a prospect CSV, a trend brief, or a set of reusable content atoms.

The concrete problem it replaces is the ad hoc prompt. Instead of asking an agent to "look at some TikToks" and getting an unstructured answer, the skills define complete research workflows with a named output format, so the agent knows which questions to ask, which data to pull, and how to present the result. The data layer is ScrapeCreators, which covers TikTok, Instagram, YouTube, Reddit, X/Twitter, LinkedIn, Facebook, Threads, Bluesky, Pinterest, Rumble, and ad libraries. Skills cover the analysis layer rather than endpoint routing alone; raw fetching requests fall to the scrapecreators-api skill, which maps them to the right endpoint and pagination guidance.

Key capabilities

• outlier-post-finder finds posts, reels, shorts, tweets, or videos that beat a creator's baseline, and returns an outlier table with repeatable patterns and hooks.
• transcript-intelligence analyzes video transcripts from TikTok, Instagram, YouTube, Facebook, X, LinkedIn, Rumble, and Reddit, producing summaries, hooks, claims, quotes, and content atoms.
• comment-mining extracts questions, objections, pain points, product ideas, and exact audience language into a voice-of-customer report.
• ad-library-teardown analyzes active Meta, Google, and LinkedIn ads for messaging angles, hooks, CTAs, offers, and test ideas.
• competitor-social-research compares competitors' social strategies and returns a brief with content pillars, gaps, and recommendations.
• influencer-prospecting builds creator prospect lists from public social data and exports a prospect CSV or table with fit score and outreach notes.
• content-repurposing converts videos, transcripts, and posts into LinkedIn posts, X threads, scripts, and newsletter or blog ideas.

Who uses it and how

• Content marketers running competitor teardowns across several brands on TikTok and Instagram to see which formats and topics work for each.
• Growth teams validating product ideas by mining social posts, comments, and Reddit threads for demand signals, pains, objections, and exact customer language.
• Influencer and partnership teams building scored prospect lists from public profile data before outreach.
• Brand and category researchers producing multi-source social listening briefs with cited examples and sentiment caveats.
• Creators and solo operators analyzing their own transcripts to pull hooks, claims, and angles they can reuse.

Getting started

Install with npx skills add ScrapeCreators/social-media-research-skills, then set export SCRAPECREATORS_API_KEY=sk_... with a key obtained from scrapecreators.com. The skills run in Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, Windsurf, VS Code, and other agents that support the Agent Skills spec.

How it compares

The facts name no comparable tools, so this project stands alone in this registry as a social media research skill set for AI agents. It should be judged against general-purpose agent prompting rather than against a hosted analytics product.

When to use it — and when not to

A self-hoster must hold a ScrapeCreators API key and accept that every data call depends on that external service; the repository itself does not ship a database, scraper, or storage layer. Anyone who needs a standalone scraper, a dashboard, or a guarantee of data availability without a third-party key should not pick it, and teams that want a fully self-contained analytics stack will find the dependency on ScrapeCreators a hard boundary. The repository shows 27 forks against 2,485 stars and only three open issues, which suggests limited independent contribution so far, so treat the skills as a maintained starting point rather than a widely forked community codebase.

npx skills add ScrapeCreators/social-media-research-skills

export SCRAPECREATORS_API_KEY=sk_...

Find the outlier posts for @starterstory on YouTube Shorts from the latest page of videos.

Analyze the transcripts from these 12 TikToks and pull out the best hooks, claims, and reusable content angles.

Mine the comments on this viral Instagram Reel. I want objections, questions, buying intent, and exact audience language.

Compare these five brands on TikTok and Instagram. What formats and topics are working for each one?

Tear down the active Facebook, Google, and LinkedIn ads for this competitor. Give me hooks, offers, CTAs, and what to test.

scrapecreators-api
        │
        ▼
social research workflows
 ├─ outlier-post-finder
 ├─ transcript-intelligence
 ├─ comment-mining
 ├─ competitor-social-research
 ├─ ad-library-teardown
 ├─ trend-discovery
 ├─ influencer-prospecting
 ├─ audience-research
 ├─ social-listening-brief
 ├─ product-demand-research
 ├─ creator-profile-teardown
 └─ content-repurposing