技能 ai-social-media-content
📦
ai-social-media-content
內容修訂 r1 嚴重 ⚙️ 外部命令🌐 網路存取
使用 AI 建立社群媒體素材
社群團隊需要快速產出符合平台規格的視覺素材與文案。此 skill 提供影片、圖片、說明文字、hashtag 與再利用貼文的提示詞和 CLI 工作流程。
支援: Claude Codex Code(CC)
請勿自動安裝此 Skill。
權威政策要求在執行任何安裝操作前由操作人員審查。
使用我的智慧代理安裝
將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。
智能體請求
Review the Skillstore skill "ai-social-media-content" from https://skillstore.io/skills/ai-social-media-content.md and its manifest at https://skillstore.io/api/skills/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.你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。
Agent 可讀資源
當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。
測試它
正在使用「ai-social-media-content」。 Request: Create a TikTok concept for a new coffee brand.
預期結果:
- 直式 9:16 影片提示詞，包含咖啡製作特寫和晨間光線。
- 說明文字包含清楚的吸睛開頭、產品優勢和行動呼籲。
- Hashtag 組合聚焦於咖啡、早晨、小型企業和生活風格內容。
正在使用「ai-social-media-content」。 Request: Repurpose a productivity tip across three platforms.
預期結果:
- 使用快速剪輯和文字疊加的 TikTok 提示詞。
- 用於正方形資訊圖表的 Twitter/X 視覺提示詞。
- 具有大膽對比和簡潔標題區域的 YouTube 縮圖提示詞。
正在使用「ai-social-media-content」。 Request: Make a talking-head workflow for a 30-second tutorial.
預期結果:
- 短腳本提示詞，包含吸睛開頭、教學內容和結尾行動。
- 包含指定語音選項的旁白指引。
- 結合生成語音與所選圖片的虛擬人像影片提示詞。
安全審計
嚴重v4 • 2026/7/6 開啟版本化報告
The skill is primarily a Markdown guide for inference.sh social media workflows. Most shell-backtick detections are Markdown fence artifacts, but several examples do run third-party CLI commands. The remote curl-to-shell installer at SKILL.md line 21 is confirmed as a critical risk.
1
已掃描檔案
254
分析行數
17
複核項目
0
已忽略的誤報
已確認的安全問題 (1)
嚴重
Pipe to shell pattern
curl -fsSL https://cli.inference.sh | sh && infsh login
The command pipes a remote installer directly into sh and then starts login. This can execute unreviewed code from the network in the user environment.
能力複核項目 (17)
這些是真實的本機能力，對此技能而言可能是預期行為，因此需要複核，但不會按已確認的惡意行為計分。
中
Ruby/shell backtick execution
```bash
The code block documents a shell workflow that installs the inference.sh CLI and runs infsh commands. This is legitimate for the skill, but it still asks users or agents to execute third-party CLI commands.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains a shell loop that calls infsh and writes carousel output files. The risk is controlled but real because it executes a third-party CLI and creates local files.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block calls infsh for image generation and includes a Twitter/X posting command. This can create external side effects if credentials are configured, so the external command risk is real.
中
Ruby/shell backtick execution
```bash
The workflow calls external model, text-to-speech, and avatar services through infsh. The commands are documented examples, but they can send script, image, and audio data to providers.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block contains infsh app run examples that call external inference providers. The arguments are mostly fixed examples, so the main risk is external execution and data egress, not shell injection.
中
Ruby/shell backtick execution
```bash
The block uses a shell variable to drive multiple infsh app runs for repurposing. No command injection is shown, but it is still real external CLI execution.
中
Ruby/shell backtick execution
```bash
The batch workflow loops over topics and runs infsh for each item while writing output files. This is a legitimate workflow with real external command and local file effects.
中
Ruby/shell backtick execution
```bash
The related-skills block uses npx skills add commands to install additional community skills. Installing code from a repository is a real external command risk and should require review.
低
Hardcoded URL
curl -fsSL https://cli.inference.sh | sh && infsh login
The URL is used in an installer command that downloads a remote script. This creates a real network dependency before local execution.
風險因素
⚙️ 外部命令 (32)
SKILL.md:20-27 SKILL.md:27-44 SKILL.md:44-49 SKILL.md:49-53 SKILL.md:53-60 SKILL.md:60-64 SKILL.md:64-69 SKILL.md:69-73 SKILL.md:73-84 SKILL.md:84-88 SKILL.md:88-105 SKILL.md:105-111 SKILL.md:111-115 SKILL.md:115-119 SKILL.md:119-123 SKILL.md:123-127 SKILL.md:127-131 SKILL.md:131-135 SKILL.md:135-139 SKILL.md:139-143 SKILL.md:143-147 SKILL.md:147-151 SKILL.md:151-156 SKILL.md:156-160 SKILL.md:160-164 SKILL.md:164-170 SKILL.md:170-188 SKILL.md:188-192 SKILL.md:192-201 SKILL.md:201-236 SKILL.md:236-251 SKILL.md:251-253
🌐 網路存取 (3)
偵測到的模式
Pipe to shell pattern
審計單位: codex 檢視審計紀錄 →
分享與引用此報告
分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。
複製報告連結
https://skillstore.io/skills/ai-social-media-content/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/ai-social-media-content?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/ai-social-media-content?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/ai-social-media-content/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/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>學術引用 (APA · BibTeX · CFF)
APA 引用
inference-sh. (2026). ai-social-media-content security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/ai-social-media-content/audits/4BibTeX 引用
@techreport{inference-sh-ai-social-media-content-2026,
author = {inference-sh},
title = {ai-social-media-content security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/ai-social-media-content/audits/4},
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 4)"
version: "unspecified"
type: report
authors:
- name: "inference-sh"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/ai-social-media-content/audits/4"
identifiers:
- type: other
value: "skillstore:ai-social-media-content:audit:4"
description: "Skillstore immutable audit report identifier"
比較變體
2 個可安裝變體每位作者的技能都保留為獨立的可安裝項目。推薦變體依 Skillstore 證據排序。
為何此變體排名第一
Skillstore 使用量較高
作者 / 來源 Skillstore 評分 證據置信度 Skillstore 使用量 已更新
ai-social-media-content
Skillstore 評分 38
證據置信度 中
Skillstore 使用量 52
已更新
2026-10-22
inference-sh-9-ai-social-media-content
Skillstore 評分 38
證據置信度 中
Skillstore 使用量 28
已更新
2026-09-09
Skillstore 評分
評分依據 證據置信度: 中 55
架構
85
可維護性
87
內容
71
社群
83
規範符合度
你可以打造什麼
推出短影音行銷活動
為 TikTok、Reels 和 Shorts 建立直式影片概念與配套說明文字。
批次建立創作者素材
從主題清單產出每週視覺素材、吸睛開頭、說明文字和旁白概念。
準備產品視覺素材
為產品發布生成產品展示提示詞、縮圖和可分享的社群圖像。
試試這些提示
建立一支短影片
為我的主題建立一個 TikTok 影片概念。包含長寬比、時長、視覺提示詞、說明文字和 hashtags。
建立輪播組合
為這個主題建立一組五張投影片的 Instagram 輪播。保持視覺風格一致，並為每張投影片包含說明文字。
跨平台再利用
將這個概念轉換成 TikTok、Twitter/X、YouTube 縮圖和 Instagram 動態消息素材。說明每個平台的格式選擇。
規劃每週內容批次
為這些主題建立七天份的內容批次。為每則貼文包含影片提示詞、說明文字、hashtags 和審核備註。
最佳實務
- 執行任何範例前，先審查安裝程式命令和 CLI 憑證。
- 提供平台、長寬比、時長、受眾、品牌風格和審核標準。
- 發布前檢查生成素材的權利、準確性、揭露需求和平台政策。
避免
- 未經驗證就執行遠端安裝程式或 npx install 命令。
- 未經明確人工核准就發布自動化社群貼文。
- 使用通用爆紅提示詞，卻沒有受眾、品牌或政策限制。
常見問題
這個 skill 需要 inference.sh 嗎？
是。這些工作流程是為 inference.sh CLI 及其託管模型應用程式撰寫的。
它可以建立影片和圖片嗎？
可以。它提供影片、圖片、縮圖、虛擬人像和語音工作流程的 CLI 範例。
它可以撰寫說明文字和 hashtag 嗎？
可以。它包含說明文字、hashtag、吸睛開頭和行動呼籲的提示詞模式。
它會直接發布到社群網路嗎？
它包含一個 Twitter/X 自動化範例，但發布需要憑證和明確核准。
安裝程式命令可以安全地自動執行嗎？
否。遠端 pipe-to-shell 安裝程式在使用前應替換或手動驗證。
哪些 AI 工具可以使用這個 skill？
報告列出支援 Claude、Codex 和 Claude Code。
開發者詳細資訊
作者
inference-sh授權
MIT
Skillstore 修訂
r1
版本提示
作者未聲明版本。
引用
c1fdca50ff516318f65fed0d7f9e82797c5171dc
維護新鮮度
2026/7/18
使用量
10 次下載 · 267 次瀏覽
檔案結構
📄 SKILL.md