AI Agent Hub
Back to skills
Social Media Content Research Assistant icon

Social Media Content Research Assistant

Content Creation Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Follow https://skillhub.cn/install/skillhub.md to install @user_825d9e7e/socialdatax-content-research-assistant.

About this skill

Problem

When planning topics, tracking competitors, or reviewing comments, operators often have to compare signals from XHS, Douyin, Kuaishou, Bilibili, Zhihu, Weibo, Video Accounts, Instagram, X, YouTube, and TikTok. Manual workflows tend to break down in three places: inconsistent platform entry points, mismatched parameter names, and weak evidence trails for comments and creator profiles. Social Media Content Research Assistant provides a read-only workflow for content operations, brand research, and creators who need structured social content evidence.

How It Works

It prefers the closest direct CLI or MCP tool for each task instead of falling back to generic keyword search. Typical paths include:
- Trends: call platform hot-list commands such as xhs hot-search, douyin hot-search, or weibo hot-search to collect topics and heat signals.
- Content search: use platform-specific search, search-videos, or search-posts commands with supported filters such as --sort-type and --publish-time-range.
- Details and comments: fetch body text, publish time, interaction metrics, comment text, and reply clues by content ID or URL.
- Creator research: retrieve profiles, verification, follower signals, and recent content lists by account ID, username, or profile URL.

Responses usually include platform, tool, arguments, and data, which helps separate raw facts, interpretations, and follow-up questions.

Boundaries

The skill is read-only: it does not log in, post, like, comment, or modify accounts. It uses the SOCIALDATAX_API_KEY environment variable at runtime, while generated skill files do not contain the key. Platform parameters and pagination differ: XHS uses --sort-type, --publish-time-range, and --note-type, while MCP tools use snake_case field names. For pagination, do not send a token on the first page, and send the full returned next_page_token unchanged for later pages. Handle balance or rate-limit messages according to the returned guidance instead of retrying aggressively.

Use Cases

  • Prepare XHS topics by checking hot lists, searching related notes, and summarizing frequent comment themes.
  • Benchmark Douyin, Kuaishou, and Weibo trends, then collect candidate creator profiles and recent content.
  • Review competitor posts by fetching Bilibili or YouTube video details, comments, and reply clues.
  • Organize cross-platform materials by separating hot-list signals, search samples, and creator content evidence.

Best For

  • Content operations staff planning XHS or Douyin topics who need hot lists, note samples, and comment feedback.
  • Brand or market researchers monitoring competitors who need cross-platform trends, creator positioning, and recent content.
  • Creators tracking overseas platforms who need post, comment, and profile evidence from X, YouTube, or TikTok.
  • Content editors compiling social assets who need to separate platform facts, interaction metrics, and follow-up angles.