agentsclimarketplace
Sign in

Content research

Skill Ootto-AI/claude-content-skills/skills/content-research

Turn Claude into a content research machine for any niche. Point it at any creator or account and it pulls their posts, finds the exact ones that blew up (the outliers), then breaks down WHY — the hook, the format, and the retention patterns — and surfaces the repeatable playbook you can copy. Use when the user says "research this niche", "what's working for @account", "find what's going viral in [niche]", "what should I post about", "study this creator", or wants a content strategy grounded in real data instead of guessing.From its SKILL.md

Install
npx -y skills add Ootto-AI/claude-content-skills --skill content-research

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 19 stars19 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.4 KB, 665 tokens by cl100k_base, as published. Nobody here has run it

Content Research — stop guessing what to post

Most people spend hours scrolling to figure out what to post. This skill turns Claude into a research machine: point it at any account or niche and it finds the exact moments that went viral and the patterns behind them — so you copy what works instead of guessing.

What it does

  1. Pull the posts. For a creator/account, gather their recent reels + view counts (Instagram Graph API via Composio, or any list of reel URLs the user provides).
  2. Find the spikes. Rank posts by views and flag the outliers — the exact posts where the account blew up vs their baseline. Those are the moments worth studying.
  3. Break down WHY. For each outlier, study the reel frame-by-frame + transcript (hand off to the reel-analyzer skill / ootto-watch) and extract the hook (first 2s), the format/structure, the pacing, and the retention pattern.
  4. Surface the playbook. Across the outliers, surface the repeatable patterns — the hook types, formats, and topics that consistently earn saves/shares — and turn them into a short, copyable plan for the user's own niche.

How to run it

  1. Ask for a creator @handle, an account URL, or a niche (+ a few example accounts).
  2. Pull their reels + view counts. Options:
    • sandcastles.ai — a research engine that pulls top channels and auto-surfaces the viral outliers + the frameworks behind them (fastest path; connect it to Claude and let it do the heavy lifting).
    • Composio Instagram tools — free/DIY: list a creator's media + insights from the Graph API.
    • Or a plain list of reel URLs the user pastes.
  3. Rank by views, compute each post's ratio vs the account median, and mark anything ~2-3x median as an outlier ("blew up here").
  4. For the top 3-5 outliers, run reel-analyzer (or ootto-watch — github.com/Ootto-AI/ootto-watch) to break down hook / format / retention.
  5. Synthesize: list the winning hook patterns, formats, and topics, then write 3 ready-to-shoot ideas for the user's own account modeled on what actually worked.

Output

A short research report:

  • The spikes — which posts blew up and by how much (vs baseline).
  • Why they worked — hook / format / retention breakdown per outlier.
  • The playbook — the repeatable patterns + 3 original ideas for your niche.

Honesty

Model the technique, never copy another creator's exact script, visuals, or caption. Cite the source reels you studied. View counts come from real data (Composio / what the user provides) — never fabricate.

Part of the Ootto content-skills marketplace. Pairs with reel-analyzer, going-viral, and reel-scripter.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most research analysis skills give in 665 tokens

Counted across 1,213 of the 2,113 authors here whose files we hold, read 2026-09-06

  • Cite sources for every important claimin 47 of 1213, across 38 files
  • Separate facts from inferences and recommendationsin 21 of 1213, across 12 files
  • Write findings to a markdown filein 19 of 1213
  • Label every insight with a confidence levelin 18 of 1213, across 8 files
  • Read product marketing context before asking questionsin 18 of 1213, across 8 files
  • Rank themes by frequency and intensityin 16 of 1213, across 6 files
  • Establish research mode before proceedingin 16 of 1213, across 6 files
  • Segment survey responses by customer tier or tenurein 16 of 1213, across 6 files
  • Categorize support tickets before analyzingin 16 of 1213, across 6 files
  • Weight research sources from the last twelve monthsin 16 of 1213, across 6 files
  • Use at least five data points per segmentin 15 of 1213, across 5 files
  • Extract verbatim quotes for all research findingsin 15 of 1213, across 5 files

Said here and by no other author read

  • Ask for a creator handle or niche
  • Gather recent posts and view counts
  • Rank posts by view count
  • Identify outliers exceeding median views
  • Analyze hook format and retention patterns
  • Synthesize repeatable patterns into a playbook

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.