小红书自动发布技能
v1.0.0自动生成并发布小红书笔记，支持用户指定主题和自动热点选题两种模式，完成内容创作和一键发布。
⭐ 0· 53·0 current·0 all-time
byXuanying Chen@luckychay
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the included scripts and docs. The skill requires the existing 'xhs' (小红书) skill and an OpenClaw agent, and the scripts call the agent to generate content and then rely on the xhs skill to publish — this is coherent for an auto-publish tool.
Instruction Scope
SKILL.md and the main script only build prompts and call 'openclaw agent --message ... --agent main' to generate content then rely on the xhs skill to publish. They do not read unrelated files or env vars. Note: the scripts instruct the agent to publish '无需确认' (without confirmation) and to '仅发布一次，无论成功与否均停止' — this reduces user-in-the-loop control and means the agent/xhs skill will perform outbound publishing automatically, which users should be aware of.
Install Mechanism
No remote download or package install is performed. The provided setup.sh operates on local files (permission changes and copying into the workspace). No URLs, extract actions, or external installers are used.
Credentials
The skill declares no required environment variables or credentials itself. It depends on an installed 'xhs' skill for publishing (which is where account credentials would live). No unrelated credentials or secrets are requested by these files.
Persistence & Privilege
always:false and user-invocable:true. The install script copies the runnable script into the workspace so it can be scheduled (cron) or invoked by the agent — this is normal. Users should note the script's default behavior to publish automatically (no confirmation) and the recommendation for cron scheduling, which can create ongoing automated publishing if configured.
Assessment
This skill appears to be what it says: a local script that builds prompts and asks your OpenClaw agent to generate posts, then relies on your installed xhs skill to publish. Before installing: 1) Confirm which Xiaohongshu account the existing xhs skill is configured to use (to avoid posting from the wrong account). 2) Be aware the script instructs the agent to publish without confirmation and to 'publish once' automatically — if you want manual review, modify the script to require confirmation before calling the xhs publish action. 3) Review and test in a safe environment (or with a test account) to verify content quality and compliance with platform rules. 4) If you plan to enable cron, confirm scheduling and frequency to avoid account-rate limits. The code is local and does not fetch remote archives or request unrelated credentials, but it will cause outbound posts via your xhs skill, so verify that skill's permissions and tokens first.Like a lobster shell, security has layers — review code before you run it.
latestvk977v0ev2h2cqpc0pjykfppy3584eka3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.