Xiaohongshu · RedNote · 小红书 · AgentSkill

Xiaohongshu Skill for AI Agents

Search notes, read detail pages, prepare publish flows, inspect selectors, and return JSON from a Python Playwright CLI that agents can call safely.

Install the Skill Command Reference Demo Outputs 
npx skills add DeliciousBuding/xiaohongshu-skill
pip install git+https://github.com/DeliciousBuding/xiaohongshu-skill.git
playwright install chromium
xiaohongshu-skill qrcode --headless=false
xiaohongshu-skill --profile brand-a search "上海咖啡" --limit=3
xiaohongshu-skill selectors --owner=publish
xiaohongshu-skill contracts --command=search
xiaohongshu-skill terminal preview with Xiaohongshu and RedNote labels
PythonPlaywright browser automation
JSONagent-readable command output
AgentSkillClaude Code, Codex, OpenClaw
MITopen source repository

Install paths

Use the path that matches your agent runner. The skills.sh and ClawHub entries point people to public install surfaces, while the pip path gives shell users the CLI directly.

npx skills add DeliciousBuding/xiaohongshu-skill
clawhub install xiaohongshu-skill
pip install git+https://github.com/DeliciousBuding/xiaohongshu-skill.git
git clone https://github.com/DeliciousBuding/xiaohongshu-skill.git <skills-dir>/xiaohongshu-skill
docker compose run --rm xiaohongshu search "美食"

What the Skill does

Search

Keyword search with sort, type, time, scope, and location filters.

Read

Fetch note details, comments, user pages, profile state, and explore feed data.

Publish

Prepare image, video, Markdown-image, and long-form posts after user approval.

Operate

Use templates, daily limits, profiles, selectors, output contracts, and SOP commands.

Agent interface

Read commands can run directly

  • search, feed, user, explore, me
  • selectors and contracts expose stable automation surfaces.
  • --profile <name> isolates sessions for separate accounts.

Write commands need approval

  • publish, comment, reply, like, and collect change account state.
  • The browser stops on captcha or verification pages.
  • Use live tests only with XHS_LIVE_TEST=1 and a test account.

LLM and search discovery

This site exposes crawlable HTML, JSON-LD, canonical links, a sitemap, and two text files for agent indexers. The terms below are visible so search engines and agent catalogs can map the project to common user prompts.

Xiaohongshu Skill RedNote Skill 小红书 AI Agent xhs browser automation AgentSkill skills.sh ClawHub Claude Code Codex OpenClaw Python Playwright

Docs for humans and agents

These links point to the project pages most likely to answer install, command, safety, and integration questions.

llms.txt llms-full.txt Install Commands Integrations Security Reference map Examples