Rising-Creator Early Detector

SKILL: Rising-Creator Early Detector
Surfaces the most-engaged recent content in a niche across TikTok (with optional Instagram and YouTube), divides engagement by each creator's follower or subscriber count, and flags those punching far above their size — cheap partnerships before prices rise. You choose the platforms, freshness window, search depth, and follower ceiling.

You are running this skill on API Direct via its MCP tools. Execute the steps below yourself by calling the named tools in order — values in <angle brackets> come from a previous step. Then deliver the result described at the end.

INPUTS:
  - niche (required): <niche — ASK THE USER>
      The content niche to scan.
  - platforms (optional): (optional — e.g. tiktok, instagram, youtube)
      Comma-separated list of platforms to run — options: tiktok, instagram, youtube. Omit to run all of them; name specific platforms to limit the run.
  - region (optional): (optional — e.g. us)
      2-letter region code to scope the scan to. Applies to the TikTok search (the only search tool with a region param).
  - freshness_days (optional): 30 (default — override if you like)
      How recent the content must be, in days. Maps directly to TikTok's publish_time (allowed values: 0, 1, 7, 30, 90, 180) and to the nearest YouTube upload window (today/this_week/this_month/this_year). Default: 30.
  - pages (optional): 2 (default — override if you like)
      How many pages of results to pull per search — more pages = a wider net of creators. Default: 2.
  - max_followers (optional): (optional — e.g. 100000)
      Ceiling on follower/subscriber count, applied in the ratio step, so only smaller, still-undervalued creators surface. Default: no cap.

PLATFORM SELECTION: some steps are gated with "Only run if {platforms} includes X". If the user supplied a `platforms` value, run only the steps whose platform is listed. If the user did NOT supply `platforms`, run ALL steps — every platform the skill supports.

STEPS:
  1. Tool `search_tiktok` — search_tiktok(query="{niche}", sort_by=most_liked, publish_time=30, pages=2)
     Base scan (always runs). Pull the most-liked TikTok videos in the niche from your freshness window. Record each author handle and the video's like count.
  2. Tool `search_tiktok_users` — search_tiktok_users(query=<tiktok_author>, pages=2)
     For each TikTok author, get follower count and video_count, then compute likes ÷ followers. Flag creators with a high ratio but a following at or below {max_followers} (no cap if unset) — rising creators to lock in early.
  3. Tool `search_instagram` — search_instagram(query="{niche}", pages=2)
     Only run if {platforms} includes instagram: pull recent niche posts on Instagram. Record each creator's username and the post's like/comment engagement.
  4. Tool `search_instagram_users` — search_instagram_users(query=<instagram_username>)
     Only run if {platforms} includes instagram: look up each creator's follower count, compute engagement ÷ followers, and flag high-ratio accounts at or below {max_followers}. (Note: search_instagram_users takes only query — no pages param.)
  5. Tool `search_youtube` — search_youtube(query="{niche}", upload_date=this_month, pages=2)
     Only run if {platforms} includes youtube: pull top recent niche videos. Record each channel name and the video's view count. Adjust upload_date to match 30: today / this_week / this_month / this_year.
  6. Tool `search_youtube_channels` — search_youtube_channels(query=<youtube_channel>, pages=2)
     Only run if {platforms} includes youtube: pull each channel's subscriber count and compute views ÷ subscribers. Then merge all platforms into one shortlist, dedupe creators that appear on more than one network, drop anyone above {max_followers}, and rank by engagement-to-follower ratio (best first), tagging each with its platform.

DELIVER: A deduped, cross-platform shortlist of undervalued creators, each tagged with its platform, follower/subscriber count, and engagement-to-follower ratio, ranked best-first for contact-worthiness.

Note: each underlying tool call is billed at its normal endpoint price; get_sentiment adds a small per-page surcharge. Page through results as needed but stop once you have enough to deliver the outcome.