How an Open‑Source AI Tool Turns Xiaohongshu Posts into Viral Hits

The open‑source Xiaohongshu MCP (Multi‑Channel Publisher) is an AI‑powered platform that assists creators in every stage of content production, from topic selection to publishing and performance analysis.

Key Features

AI‑driven topic prediction : Uses recent 30‑day hot‑list data and user‑preference algorithms to suggest high‑traffic, low‑competition topics, showing estimated views and follower potential.

Automatic copy, title and tag generation : Generates three stylistic versions of copy, titles with a pain‑point + solution format, and matches 10+ precise tags.

Data monitoring & competitor analysis : Tracks views, likes, saves, follower growth; visualizes trends; allows input of competitor accounts to dissect their topic, copy structure and posting time.

Batch account management & scheduled publishing : Supports simultaneous handling of 5+ accounts, scheduled posting, bulk draft saving and cross‑platform sync.

Open‑source and customizable : Full source code on GitHub, can be self‑hosted, customize copy style or tag library, no commercial licensing.

Typical Use Cases

1. New creator scaling from zero

Use AI topic prediction to find a niche (e.g., affordable makeup).

Generate three copy drafts, select preferred style, and apply auto‑matched tags.

Schedule posting at AI‑recommended peak hour and monitor real‑time metrics.

Apply competitor analysis to replicate successful patterns, resulting in rapid follower growth.

2. Product‑review (affordable goods) influencer

Import a list of products to promote.

AI creates tailored copy and titles for each product, suggesting relevant tags and cover ideas.

Schedule releases and adjust promotion based on “likes‑to‑commission conversion” data, tripling earnings.

3. Reviving an old account

Upload a low‑performing note; AI flags weak title, generic tags, and disorganized copy.

Implement AI suggestions: rewrite title with a clear hook, replace tags with niche‑specific ones, restructure copy into clear points.

After republishing, views jump from hundreds to tens of thousands and engagement rises sharply.

Quick Start Guide

Step 1: Deploy the tool

Two options are provided.

Local deployment (recommended for data privacy) :

# Clone the repository
git clone https://github.com/xpzouying/xiaohongshu-mcp.git
cd xiaohongshu-mcp
# Install dependencies
pip install -r requirements.txt
# Copy example env file and fill in Xiaohongshu Cookie
cp .env.example .env
# Run the application
python main.py

Access the UI at http://localhost:8501.

Online demo : Use the demo link in the README to try core features without installation.

Step 2: Generate a note

Click “New Note” and fill in the domain, topic direction, and desired style (e.g., lifestyle, tutorial, recommendation).

Press “Generate Content”; the AI returns a title, copy and a set of tags within seconds.

Preview and optionally regenerate or manually tweak the output.

Step 3: Publish and optimize

Schedule the post for an AI‑suggested active period.

Monitor metrics such as collection rate and comment interaction to gauge content value.

Use the competitor‑analysis feature to discover winning topics and posting times, then iterate.

Final Thoughts

The MCP does not replace a creator’s personal voice; it removes repetitive trial‑and‑error by handling topic selection, copy drafting and tag optimization, allowing creators to focus on adding authentic experiences and refining content quality.

Being open‑source, the project continues to add features like AI‑generated cover copy, compliance risk detection, and upcoming fan‑need analysis, making it increasingly aligned with real‑world Xiaohongshu operations.

Project repository: https://github.com/xpzouying/xiaohongshu-mcp