Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Automated Post

v1.0.0

通过调度文案与配图代理,实现图文帖子发布的确认或自动化两种全流程操作与归档管理。

0· 34·1 current·1 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for yifeiwang1981/automated-post.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Automated Post" (yifeiwang1981/automated-post) from ClawHub.
Skill page: https://clawhub.ai/yifeiwang1981/automated-post
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install automated-post

ClawHub CLI

Package manager switcher

npx clawhub@latest install automated-post
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
The stated purpose (coordinate copywriting and image-generation agents and publish/archive posts) is coherent with the SKILL.md. However, the instructions assume existence of other local skills/scripts (post_skill.js, append_doc.js) and a Feishu integration without declaring required credentials or explaining which website is targeted.
!
Instruction Scope
SKILL.md instructs the agent to run 'openclaw agent' for other agents, cd into specific home-directory paths (~/.openclaw/... and ~/.openclaw/workspace-xichun/...), and execute local node scripts and a feishu_doc tool. Those steps access local filesystem and external services and require tokens, but the skill does not declare or justify access to those secrets or files.
Install Mechanism
There is no install spec (instruction-only skill), so nothing is written to disk by an installer. The runtime relies on existing binaries/tools (node, openclaw CLI, feishu_doc) being present.
!
Credentials
The SKILL.md refers to a 'doc_token', Feishu scopes (docs:document:create, docs:upload, etc.), and posting to a website via a local post_skill.js, but requires.env lists none. Required credentials (Feishu tokens, website API keys) are missing from the declared requirements — an inconsistency that could lead the agent to attempt to read tokens from other places or fail silently.
Persistence & Privilege
always:false (no forced inclusion) and autonomous invocation is allowed (platform default). The skill will instruct running other agents and local scripts which increases its operational scope but it does not request persistent/system-wide privileges itself.
What to consider before installing
Before installing: 1) Verify the exact credentials this skill needs (Feishu doc tokens, website API keys) — the SKILL.md references them but the skill does not declare them. 2) Inspect the local scripts it calls (~/.openclaw/workspace/skills/post_to_website/post_skill.js and ~/.openclaw/.../append_doc.js) to confirm what they do and where they get credentials. 3) Ensure feishu_doc and node are installed and that any tokens are stored securely (not in plain files). 4) If you don't already have the referenced workspaces/scripts, ask the author for full code and a clear list of required env vars and permissions before using. 5) Consider running in confirmation mode (not fully automated) or restricting agent autonomy until you validate the toolchain.

Like a lobster shell, security has layers — review code before you run it.

latestvk975zencaa4ay79t1rtwydxnyn85maqr
34downloads
0stars
1versions
Updated 8h ago
v1.0.0
MIT-0

SKILL.md - 袭人图文发布自动化技能

技能名称

automated_post - 袭人图文发布自动化

技能描述

袭人通过调度黛玉(文案)和惜春(配图)完成图文帖子发布的完整自动化流程。支持两种模式:确认模式和自动化模式。

两种模式

模式1:确认模式(每步确认)

每执行一步都要等非哥确认后再执行下一步。

模式2:自动化模式(中间不确认,最终确认)

中间过程自动执行,只在最终发布前让非哥确认文案和图片。


执行流程

步骤1:制定方案

收到非哥的主题后,制定执行方案,列出步骤和负责人(袭人、黛玉、惜春)。

步骤2:执行方案

根据选择的模式执行:

确认模式流程:

第1步:黛玉写文案 → 非哥确认
第2步:提取关键词 → 非哥确认
第3步:惜春生成图片 → 非哥确认
第4步:发布到网站

自动化模式流程:

第1步:黛玉写文案(自动执行)
第2步:提取关键词(自动执行)
第3步:惜春生成图片(自动执行)
→ 返回中间结果给非哥
非哥确认后:
第4步:发布到网站

Agent启动命令

启动黛玉写文案

openclaw agent --agent daiyu --message "<主题和要求>" --timeout 120

启动惜春生成图片

openclaw agent --agent xichun --message "<图片提示词>" --timeout 180

发布到网站

cd ~/.openclaw/workspace/skills/post_to_website
node post_skill.js "<作者>" "<文案>" "<标签>" "<图片路径>"

时间记录规范

每次执行必须记录时间:

步骤内容开始时间结束时间耗时
开始收到主题HH:MM:SS--
第1步黛玉写文案HH:MM:SSHH:MM:SSX秒
第2步提取关键词HH:MM:SSHH:MM:SSX秒
第3步惜春生成图片HH:MM:SSHH:MM:SSX秒
完成等待非哥确认HH:MM:SS--

注意

  • 即使某步骤耗时0秒也要记录
  • 计算实际工作耗时(不含等待确认时间)
  • 每次发布后汇总时间给非哥

袭人卡通形象固定提示词

生成袭人卡通形象时使用:

可爱卡通形象的女孩,古典中国风汉服打扮,大眼睛甜美笑容,活泼可爱

图片路径规范

惜春生成的图片保存在:

~/.openclaw/workspace-xichun/skills/hima_painter/images/

素材归档(飞书文档)

每次发布后自动归档到飞书文档。

归档规则

  • 每天一个文档:文档名称格式 YYYY-MM-DD 鸿蒙智行发帖记录
  • 第一篇帖子:创建当天新文档
  • 后续帖子:追加到已有文档

实现方式

使用 feishu_doc 工具:

  1. 查询当天是否有对应日期的文档
  2. 如果没有,用 feishu_doc create 创建新文档
  3. 如果有,用 feishu_doc append 追加内容
  4. 用 feishu_doc upload_image 上传图片

脚本工具

# 追加内容到当天文档
node ~/.openclaw/workspace/skills/feishu-send-image/scripts/append_doc.js <文档token> <内容>

# 上传图片到文档
feishu_doc action=upload_image doc_token=<文档token> file_path=<图片路径>

需要权限

  • docs:document:create
  • docx:document
  • docs:document.media:upload

使用示例

示例1:确认模式

非哥:袭人上学记
袭人:制定方案 → 确认模式
第1步完成 → 非哥确认 → 第2步 → ...

示例2:自动化模式

非哥:袭人放学回家
袭人:制定方案 → 自动化模式
执行第1-3步 → 返回结果 → 非哥确认 → 发布

技能创建时间:2026-03-16 作者:袭人

Comments

Loading comments...