技能 ai-social-media-content
📦

ai-social-media-content

内容修订 r1 严重 ⚙️ 外部命令🌐 网络访问

创建 AI 社交媒体内容

社交媒体团队需要稳定地产出适合各平台发布的创意、文案、图片和视频。此技能提供结构化的 inference.sh 工作流,用于创建和再利用社交内容。

支持: Claude Codex Code(CC)
⚠️ 38 差

此技能属于一个技能包

一条命令安装完整技能包,获得任务所需的全部技能。

使用我的智能体安装

将此请求复制给你的智能体,其中包含权威 Skill 页面和清单。

智能体请求
Review the Skillstore skill "ai-social-media-content" from https://skillstore.io/skills/inference-sh-9-ai-social-media-content.md and its manifest at https://skillstore.io/api/skills/inference-sh-9-ai-social-media-content/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.

你的智能体仍应展示其计划,并按安全策略请求必要确认。

测试它

正在使用“ai-social-media-content”。 需要一个关于两分钟效率法则的 TikTok 创意。

预期结果:

一个 30 秒竖屏视频概念,包含快速开头钩子、三个场景节奏、简洁文案和实用标签。

正在使用“ai-social-media-content”。 将一款咖啡产品发布内容再利用到三个平台。

预期结果:

  • 一个 Reels 产品展示概念,采用暖色灯光和开箱序列。
  • 一个涵盖卖点、成分和使用场景的轮播图大纲。
  • 一个带有简洁发布信息的 Twitter/X 视觉帖子。

正在使用“ai-social-media-content”。 为一个关于 AI 趋势的视频创建缩略图创意。

预期结果:

一个高对比度缩略图方向,包含清晰的焦点主体、易读的文字区域和匹配的文案角度。

安全审计

严重
v5 • 2026/7/9 打开版本化报告

Most static external-command hits were Markdown fence noise or intended inference.sh examples, but several real risks remain. The skill includes a critical curl-to-shell installer, many external cloud CLI calls, remote skill installation examples, and a Twitter/X posting workflow that requires explicit user approval.

1
已扫描文件
251
分析行数
17
复核项
0
已忽略误报

已确认安全问题 (2)

严重
Pipe to shell pattern
curl -fsSL https://cli.inference.sh | sh && infsh login
The command uses curl to download an installer and pipes it directly to sh. This executes remote code without local review and should block automatic publication.
中
Automated Social Media Posting Workflow
The skill includes a Twitter/X workflow that posts text and media through twitter/post-tweet. This can publish content externally if an agent runs it without explicit user approval.
Lines 76-80 explicitly label Twitter automation and call a posting app with post text and media. The risk is contextual because posting depends on account authorization and user intent.
能力复核项 (17)

这些是真实的本地能力,对此技能可能是预期行为,因此需要复核,但不会按已确认恶意行为计分。

严重
Ruby/shell backtick execution
```bash
The range contains a shell example that fetches an installer with curl and pipes it to sh before login. Executing a remote script without inspection is a real code execution risk.
严重
Hardcoded URL
curl -fsSL https://cli.inference.sh | sh && infsh login
The URL is fetched by curl and piped directly into sh, causing remote content to execute locally. This is a high-impact installation pattern even if the service is legitimate.
高
Ruby/shell backtick execution
```bash
The related skills section uses npx skills add commands to install additional remote skills. Installing more skills from an external source is a supply-chain and permission expansion risk.
中
Ruby/shell backtick execution
```bash
The code block runs infsh app run against a third-party video model and sends prompt content externally. This is intended behavior, but it is real external command and network activity.
中
Ruby/shell backtick execution
```bash
The carousel workflow uses a shell loop to call infsh repeatedly and write JSON output files. This is legitimate functionality, but it is real external command execution with local file output.
中
Ruby/shell backtick execution
```bash
The thumbnail workflow invokes infsh app run against an external image model. The command is expected for the skill, but it sends prompt data to a third-party service.
中
Ruby/shell backtick execution
```bash
The range contains infsh commands that generate a visual and call twitter/post-tweet with text and media. This can publish externally through a connected account.
中
Ruby/shell backtick execution
```bash
The talking head workflow calls external text, speech, and avatar generation apps and writes outputs to files. This is intended, but it transmits content and asset URLs to third-party services.
中
Ruby/shell backtick execution
```bash
The example runs infsh app run for an external video generation model. The command is purposeful, but it is real external command execution and data transfer.
中
Ruby/shell backtick execution
```bash
The tutorial workflow invokes infsh app run for external video generation. This is legitimate for the skill, but it sends prompt content to a third-party service.
中
Ruby/shell backtick execution
```bash
The product showcase workflow calls an external video generation app through infsh. It is intended use, but it performs external command and network activity.
中
Ruby/shell backtick execution
```bash
The lifestyle example invokes infsh app run against an external video model. This sends prompt content outside the local environment.
中
Ruby/shell backtick execution
```bash
The behind-the-scenes example calls infsh app run for an external video model. The text is benign, but the command is still external execution with network transfer.
中
Ruby/shell backtick execution
```bash
The caption workflow runs infsh against an OpenRouter Claude model and sends prompt text to an external service. This is real external command and network behavior.
中
Ruby/shell backtick execution
```bash
The hook generation example runs infsh against an external text model. It is expected for the skill, but it transfers prompt content to a third-party service.
中
Ruby/shell backtick execution
```bash
The multi-platform workflow defines a shell variable and uses it in multiple infsh commands. It is a real shell workflow that sends content to external generation services.
中
Ruby/shell backtick execution
```bash
The batch workflow loops over topics, calls infsh repeatedly, and writes output files. This is real external command execution and local file creation.

检测到的模式

Pipe to shell pattern
审计者: codex 查看审计历史 →
分享与引用此报告

分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据,不判断该 Skill 是否安全。

打开版本化报告
安全评估

复制报告链接

https://skillstore.io/skills/inference-sh-9-ai-social-media-content/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/inference-sh-9-ai-social-media-content/security.svg)](https://skillstore.io/skills/inference-sh-9-ai-social-media-content?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/inference-sh-9-ai-social-media-content?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/inference-sh-9-ai-social-media-content/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/inference-sh-9-ai-social-media-content.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
学术引用 (APA · BibTeX · CFF)

APA 引用

inference-sh-9. (2026). ai-social-media-content security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/inference-sh-9-ai-social-media-content/audits/5

BibTeX 引用

@techreport{inference-sh-9-inference-sh-9-ai-social-media-content-2026, author = {inference-sh-9}, title = {ai-social-media-content security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/inference-sh-9-ai-social-media-content/audits/5}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "ai-social-media-content security audit report (audit version 5)" version: "unspecified" type: report authors: - name: "inference-sh-9" date-released: "2026-07-09" url: "https://skillstore.io/skills/inference-sh-9-ai-social-media-content/audits/5" identifiers: - type: other value: "skillstore:inference-sh-9-ai-social-media-content:audit:5" description: "Skillstore immutable audit report identifier"

比较变体

2 个可安装变体

每位作者的技能都保持为独立的可安装项。推荐变体按 Skillstore 证据排序。

为什么此变体排名第一

Skillstore 使用量更高
inference-sh 推荐

ai-social-media-content

Skillstore 评分 38
证据置信度 中
Skillstore 使用量 47
已更新

2026-09-09

inference-sh-9-ai-social-media-content

Skillstore 评分 38
证据置信度 中
Skillstore 使用量 25
已更新

2026-09-09

Skillstore 评分

评分依据 证据置信度: 中
55
架构
85
可维护性
87
内容
69
社区
83
规范符合性

你能构建什么

规划短视频

为 TikTok、Reels 或 Shorts 创建竖屏视频概念、提示词、文案和标签。

再利用营销活动素材

将一个营销活动创意转化为适用于多个渠道的图片、视频、缩略图和文案格式。

构建产品发布内容

以一致的品牌风格生成产品展示提示词、轮播图概念和发布文案。

试试这些提示

创建一条帖子
为我的主题创建一个 Instagram Reel 概念。包括开头钩子、视觉方向、文案和十个相关标签。
起草视频脚本
为我的受众写一个 30 秒 TikTok 脚本。包括场景节奏、旁白文本、屏幕文字和结尾行动号召。
跨平台再利用
将这个概念再利用为 TikTok 视频、Instagram 轮播图、YouTube 缩略图和 Twitter/X 视觉帖子。保持信息一致。
构建每周批量内容
为这个品牌创建一个五天社交内容计划。包括每天的主题、素材类型、视觉提示词、文案和审核备注。

最佳实践

  • 发布前审查生成帖子的准确性、权利和品牌语气。
  • 在每次请求中提供平台、受众、语气、产品详情和视觉风格。
  • 在任何工作流发布内容或发帖到外部账号之前,使用人工审批。

避免

  • 不要在未进行事实核查和必要法律审查的情况下发布生成的声明。
  • 不要在提示词中包含私密客户数据、未发布计划或凭据。
  • 不要在未验证命令和目标位置的情况下运行安装程序或发帖命令。

常见问题

此技能面向哪些平台?
它涵盖 TikTok、Instagram Reels、Instagram 信息流帖子、YouTube Shorts、YouTube 缩略图和 Twitter/X 视觉内容。
它会直接创建图片和视频吗?
它提供 inference.sh CLI 工作流,用于调用外部图片、视频、语音和文本生成模型。
它能帮助生成文案和标签吗?
可以。它包含文案、开头钩子、行动号召和标签生成工作流。
此技能会自动发布帖子吗?
它包含一个 Twitter/X 发帖示例。用户应在任何内容发布前要求明确确认。
需要哪些设置?
用户需要 inference.sh CLI、账号、身份验证,以及对所选模型或应用的访问权限。
生成的内容可以直接发布吗?
不是。用户应在发布前审查准确性、知识产权、安全性和平台政策合规性。

开发者详情

许可证

MIT

Skillstore 修订

r1

版本提示

作者未声明版本。

引用

3e4b6c31a74a3bd1a291c98cf585d720cb9fbc88

维护新鲜度

2026/7/18

使用量

17 次下载 · 119 次浏览

文件结构

📄 SKILL.md

inference-sh-9 的更多技能

查看全部
查看全部