Token导航 LogoToken导航TokenDH.com
效率操作浏览器clawhub未标认证来源可访问clear审计提醒

rednote-ops红记操作

Agent Skill

rednote-ops 用于整理文档、README、Markdown 和说明材料,适合在 OpenClaw 中需要把零散信息整理成结构清晰的文档时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,525

周安装

144

GitHub Stars

公开资料未说明

下载量

1,129
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:rednote-ops(红记操作)
来源仓库:https://github.com/spencer17x/rednote-ops
安装命令:
openclaw skills install rednote-ops
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install rednote-ops

简介

端到端小红书操作技巧,用于统一内容风格与配置文件管理。

  • 支持按预设风格生成帖子并选择发布渠道。
  • 提升跨平台内容一致性与管理效率。适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。
  • 需提前定义操作模板与风格参数。rednote-ops 属于效率类 Skill,可作为该场景下的辅助能力补充。
  • 建议结合 rednote 主技能协同使用。

SKILL.md

name
xiaohongshu-ops
description
End-to-end Xiaohongshu (小红书 / RedNote) operations skill for setting content style, saving an operating profile, generating posts in that house style, choosing or generating suitable images, and publishing through the dedicated independent browser flow. Use when the user says things like “帮我运营小红书”, wants a Xiaohongshu operating style configured, wants ongoing post generation in a fixed house style, or wants to test or run a full Xiaohongshu publishing workflow.

Xiaohongshu Ops

Use this skill as the single entry point for Xiaohongshu operations.

Workflow

  1. Read data/xiaohongshu/profile.json when it exists.
  2. If profile setup is incomplete, ask concise setup questions and save the result with scripts/init_profile.js.
  3. Read data/xiaohongshu/rules.md for current local operating rules.
  4. Generate topic/title/body/page structure in the configured house style.
  5. Choose image strategy from stock, cards-lite, or mixed.
  6. Build assets and publish through the dedicated independent browser flow when publishing is requested.

Profile setup

Keep setup questions short. Cover positioning, goals, audience, style, image mode, publish preference, and whether auto learning should be enabled. Do not ask again for information the user already provided.

Watermark rule

Do not remove watermarks. Do not instruct or script watermark removal. If an image source has watermarks, reject it and choose a different source.

Auto learning

Support automatic daily learning in reference mode. If autoLearning.enabled is true in the profile, treat trend/style learning as a daily background reference task. Write learning outputs into local reference files instead of silently mutating the main profile.

Publish path

Use the dedicated independent browser profile for real publishing:

  • ~/.openclaw/browser/xhs-independent-user-data

Do not treat the user's normal browser or OpenClaw's shared browser profile as the canonical publish path.

Offer first-run publish test

After setup, ask the user whether to do a first publish test.

Recommended wording:

  • 是否要先自动生成并准备发布一篇测试笔记,顺手验证小红书发布链路是否正常?

References

Read these only when needed:

  • references/questions.md — profile intake prompts
  • references/profile-schema.json — profile fields and expected structure
  • references/workflow.md — fuller operating workflow
  • references/content-rules.md — title/body constraints and writing rules
  • references/content-templates.md — content structure patterns
  • references/content-output-format.md — expected output shape
  • references/content-ideas.md — topic/angle inspiration
  • references/image-sources.md — image sourcing guidance
  • references/stock-providers.md — stock image source notes
  • references/cards-usage.md — card-generation guidance
  • references/cards-input-example.json — card input example
  • references/publish-independent-browser.md — dedicated browser publish path
  • references/publish-usage.md — publish flow details
  • references/auto-learning.md — daily reference-learning behavior
  • references/repo-notes.md — repository-oriented notes for standalone Git usage

Scripts

Use these scripts as low-freedom helpers when deterministic execution is better than rewriting logic inline:

  • scripts/init_profile.js
  • scripts/fetch_stock_images.js
  • scripts/select_image_mode.js
  • scripts/build_image_pipeline.js
  • scripts/run_ops_once.js
  • scripts/build_xhs_post.js
  • scripts/generate_xhs_cards.js
  • scripts/xhs_publish.js
  • scripts/xhs_independent_browser.js
  • scripts/xhs_independent_publish.js

Safety rule

Default to the profile's publish preference. If the profile says review-before-publish, stop for review. If the profile allows direct publish and the user asks for publishing, use the independent browser path.

File convention

Persistent config:

  • data/xiaohongshu/profile.json

Runtime outputs can go under:

  • /tmp/openclaw/uploads/xhs_*

Content storage rule

Do not treat generated post content as a long-term local content library. For each new run, default to regenerating:

  • title
  • body
  • page structure
  • images

Use runtime outputs as temporary working files only. Do not rely on old note content by default unless the user explicitly asks to reuse an older draft.

Active content constraints

Read references/content-rules.md when generating Xiaohongshu copy. Read data/xiaohongshu/rules.md as the local source of truth for current Xiaohongshu operating rules. Current default rules:

  • title must be a natural phrase within 20 characters
  • mixed Chinese/English titles should preserve important English terms instead of clipping them mid-word
  • body should stay within 2000 characters

Integration

This skill orchestrates:

  • content generation
  • image strategy selection
  • stock-image planning
  • cards-lite fallback when needed
  • dedicated-browser publishing

Use the saved profile as the style source of truth.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

72.82%
按下载量换算822

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills