Search

Xiaohongshu Creator Research with AI Agents

Let an AI agent run full Xiaohongshu creator research in one flow — search candidate accounts by niche, pull creator profiles, scan note lists and content style, then return a comparison table you can use to pick influencers for a campaign.

Xiaohongshu Creator Research with AI Agents

Tool names in this guide are for reference only and may have changed or been renamed. Please use the latest tools in the tool marketplace when implementing.

Key takeaways

  • Search Xiaohongshu candidate creators by niche from one plain-English prompt
  • Judge fit from profile and note content, not just follower count
  • The agent returns a creator comparison table so campaign picks are evidence-based

Tools used

Why picking creators is the real grind

On Xiaohongshu, the time sink in influencer marketing isn't writing the brief — it's choosing who to work with. You search a niche, open account after account, check follower counts, scan recent notes, judge whether the tone fits, and spot whether the feed is wall-to-wall sponsored posts. A dozen candidates in, half a day is gone and all you have is scattered screenshots and notes you can't compare side by side. Worse, follower count isn't fit: a 500K general-lifestyle account may convert worse for your vertical product than a sharp 30K niche creator.

An AI agent collapses that "search — profile — read content" loop. Instead of tapping through the app, you describe the niche and your requirements in one sentence, and the agent calls a Xiaohongshu creator-search tool (such as ae_xiaohongshu_user_search), a profile tool (such as ae_xiaohongshu_user_get_profile), and a note-list tool (such as ae_xiaohongshu_user_note_list), then merges everything into a comparison table.

The workflow in four steps

  1. Define the niche — give the agent a niche or keywords plus your filtering criteria.
  2. Search candidates — pull a batch of matching creator accounts by keyword.
  3. Read the profile — fetch each candidate's profile for follower scale, positioning and baseline data.
  4. Scan the content — go through each creator's note list to judge content quality, posting cadence and ad density, then roll it all up into a comparison table.

How to run each step

Step 1 — Describe the niche

Tell the agent plainly: use ae to research Xiaohongshu creators around example niche (say, home kitchen gadgets) — find a batch of candidates and compare their profiles and content style. The agent interprets this and picks the matching tools. No API wiring, no login.

Step 2 — Search candidate creators

The agent calls a creator-search tool (such as ae_xiaohongshu_user_search) to return a batch of matching accounts for your keywords. You can hand it several keywords at once so it pulls accounts across different angles of the same niche instead of searching term by term.

Step 3 — Pull creator profiles

For each candidate, the agent calls a profile tool (such as ae_xiaohongshu_user_get_profile) to get follower scale, account positioning, verification and other baseline data. This turns "looks relevant" into "comparable on data" and is the foundation for the comparison table.

Step 4 — Scan notes, build the table

Next it calls a note-list tool (such as ae_xiaohongshu_user_note_list) to go through each creator's recent posts and judge content quality, posting cadence, tone and how much is sponsored. The agent merges profile and content signals and returns a creator comparison table — who is tightly on-niche, who posts consistently, who runs too many ads, all at a glance.

Follower count only vs. profile + content

Dimension Picking by follower count Picking by profile + content
Fit Pick whoever has reach; easy to grab a general account Match on niche focus and tone
Content quality Invisible Read the note list directly
Ad risk You learn it after the campaign See the sponsored ratio in past notes
Comparison Scattered screenshots, hard to align A consistent comparison table
Speed Tap through one by one, half a day per batch One prompt per batch

How to get started

  1. Sign up — new users get 1,000 free credits.
  2. Connect in-app inside an agent like WorkBuddy or Trae, or paste a one-line MCP config into Claude Code, Cursor or VSCode.
  3. Give the instruction in natural language and let the agent search, profile, scan and summarize into a table.

FAQ

Can it find creators by niche? Yes. Hand the agent a niche or keywords and it uses a creator-search tool to return matching accounts; you can pass several keywords at once to cover different angles.

Can it tell a creator's content style? Yes. The agent uses a note-list tool to go through a creator's recent posts and judge niche focus, posting cadence and ad density — not just follower count.

How fresh is the data? The agent calls the tools live on each run, so you get the current profile and note list rather than a stale saved snapshot. See the current coverage on the tools page.


Sign up · Browse the tools · Connect your agent