📦
video-sync
89Sync Video With Shot Analysis
由 eternityspring
Shot notes are difficult to compare with footage when they live in separate files. This skill combines both into a synchronized video with an animated shot panel.
设计与创意 查看
技能 native-subtitle-quote-image
📦
native-subtitle-quote-image
内容修订 r1 中风险 ⚡ 包含脚本⚙️ 外部命令🌐 网络访问🔑 环境变量📁 文件系统访问
Create Native Subtitle Quote Images
Turning selected video moments into readable subtitle graphics can require careful timing, cropping, and review. This skill creates compact 3:4 quote collages from authorized local videos or online sources.
支持: Claude Codex Code(CC)
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 "native-subtitle-quote-image" from https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image.md and its manifest at https://skillstore.io/api/skills/chengyi-ai-native-subtitle-quote-image/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.你的智能体仍应展示其计划，并按安全策略请求必要确认。
Agent 可读资源
当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。
测试它
正在使用“native-subtitle-quote-image”。 A local interview video with five reviewed timestamps and burned-in Chinese subtitles.
预期结果:
A 3:4 JPG collage set that preserves the original subtitles, plus a contact sheet and timestamp record.
正在使用“native-subtitle-quote-image”。 A lecture video without burned-in subtitles and five reviewed lines with timestamps.
预期结果:
A 3:4 collage with clearly labeled scripted subtitles drawn over real frames, plus a reviewable line record.
正在使用“native-subtitle-quote-image”。 An authorized YouTube link that requires an approved browser session.
预期结果:
A consent-gated download and a checked subtitle collage, or a clear stop report when access remains unavailable.
安全审计
中风险v1 • 2026/9/9 打开版本化报告
The static findings are false positives for documented, bounded media processing, fixed update checks, normal file handling, and standard Unicode ranges. A medium residual risk remains because the YouTube recovery workflow can read browser session cookies after user authorization.
10
已扫描文件
2,140
分析行数
0
复核项
0
已忽略误报
已确认安全问题 (1)
中
Sensitive Browser Session Access
The YouTube recovery workflow can ask yt-dlp to read Chrome session cookies for a private or verification-gated video. Authorization and no-export rules reduce risk, but a browser session remains sensitive account material.
The cited workflow explicitly requests permission to read Chrome cookies and then passes them to yt-dlp. The same text prohibits exporting them, so the risk is residual sensitive access rather than demonstrated exfiltration.
风险因素
⚙️ 外部命令 (42)
scripts/check_environment.py:31 scripts/check_update.py:120 scripts/native_subtitle_stitch.py:29 scripts/native_subtitle_stitch.py:50 SKILL.md:14-16 SKILL.md:16-18 SKILL.md:18-19 SKILL.md:19-20 SKILL.md:20 SKILL.md:27-28 SKILL.md:28 SKILL.md:37-44 SKILL.md:44 SKILL.md:46-55 SKILL.md:55-57 SKILL.md:57 SKILL.md:59-61 SKILL.md:61-70 SKILL.md:70-71 SKILL.md:71-82 SKILL.md:82-85 SKILL.md:85-89 SKILL.md:89-93 SKILL.md:93-103 SKILL.md:103 SKILL.md:105-108 SKILL.md:108-112 SKILL.md:112-121 SKILL.md:121-123 SKILL.md:123-128 SKILL.md:128-130 SKILL.md:130 SKILL.md:136-146 SKILL.md:146-148 SKILL.md:148-149 SKILL.md:149-155 SKILL.md:155-159 SKILL.md:159-161 SKILL.md:161 SKILL.md:174 SKILL.md:178-195 SKILL.md:195
🌐 网络访问 (5)
审计者: codex
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。
复制报告链接
https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/chengyi-ai-native-subtitle-quote-image/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/chengyi-ai-native-subtitle-quote-image.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
chengyi-ai. (2026). native-subtitle-quote-image security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image/audits/1BibTeX 引用
@techreport{chengyi-ai-chengyi-ai-native-subtitle-quote-image-2026,
author = {chengyi-ai},
title = {native-subtitle-quote-image security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image/audits/1},
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: "native-subtitle-quote-image security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "chengyi-ai"
date-released: "2026-09-09"
url: "https://skillstore.io/skills/chengyi-ai-native-subtitle-quote-image/audits/1"
identifiers:
- type: other
value: "skillstore:chengyi-ai-native-subtitle-quote-image:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 评分
评分依据 证据置信度: 中 68
架构
85
可维护性
87
内容
65
社区
83
规范符合性
你能构建什么
Build Quote Posts
Create a compact subtitle collage from an authorized interview or talk for social publishing.
Prepare Research Notes
Capture verified moments from a lecture or recorded meeting with timestamps and readable subtitles.
Package Video Highlights
Turn approved video excerpts into reviewed image assets for an editorial or marketing workflow.
试试这些提示
Make a Basic Collage
Use this skill to create a 3:4 subtitle collage from my authorized local video. Keep the original burned-in subtitles and review every image.
Use Selected Moments
Create a native subtitle collage from this video using these five timestamps. Sample nearby frames, keep the subtitle band complete, and return the checked files.
Render Reviewed Lines
Create a scripted subtitle collage from this authorized video and my reviewed timestamps and lines. Label the result as scripted subtitles and check readability.
Process an Authorized URL
Process this authorized video URL into a 3:4 quote collage. Check access and subtitle type first, request consent before browser-cookie recovery, then render and verify each output.
最佳实践
- Confirm source rights, subtitle type, language, duration, and target image count before rendering.
- Use reviewed timestamps and inspect both the contact sheet and every full-size JPG.
- Keep native and scripted subtitle modes separate, and label the selected mode in delivery.
避免
- Do not use native mode to translate, rewrite, OCR, or redraw burned-in subtitles.
- Do not claim scripted text is original on-screen subtitle content.
- Do not bypass access controls or read browser cookies without explicit user consent.
常见问题
What video sources are supported?
The skill supports local videos and authorized online video URLs, subject to access and tool availability.
What is native subtitle mode?
It crops and arranges pixels that already contain burned-in subtitles. It does not alter the subtitle text.
What is scripted subtitle mode?
It draws reviewed lines onto real video frames when burned-in subtitles are unavailable or unsuitable.
Can it improve a low-resolution video?
It can create a larger layout, but it cannot restore detail that is absent from the source.
Does it download or install dependencies?
It checks dependencies and reports missing tools. Installation requires user approval and is not performed silently.
How are browser cookies handled?
Cookie access is used only after a failed public request and explicit consent. Cookies must not be exported, logged, or stored.
开发者详情
作者
chengyi-ai许可证
MIT
Skillstore 修订
r1
版本提示
作者未声明版本。
引用
dd2fcaef932574591f2502a46fb2362cde6fa9a1
维护新鲜度
2026/9/9
使用量
0 次下载 · 0 次浏览
相关技能
查看全部📦
byted-mediakit-editing
80Edit Audio, Video, and Image Media
由 volcengine
Media editing tasks often require many specialized commands and careful parameter choices. This skill routes requests to the right MediaKit operation and guides reliable media assembly.
设计与创意 查看
📦
ponyflash
78使用 PonyFlash 生成和编辑 AI 媒体
由 ponyflash
AI 媒体工作通常需要分别使用生成、编辑和字幕工具。PonyFlash 将 SDK 指南与 FFmpeg 工作流结合起来，支持图像、视频、语音、音乐和导出。
设计与创意 查看
📦
video-inpainting
78使用 RunComfy 编辑视频区域
由 agentspace-so
跨移动视频帧的区域编辑很难稳定地选择处理路径并编写一致的提示词。此技能会选择一条 RunComfy 路径，并为清理或替换提供安全的提示词模式。
设计与创意 查看
📦
hyperframes-studio
77Build Readable HyperFrames Timelines
由 heygen-com
HyperFrames projects can become difficult to edit when scenes and media create unlabeled timeline rows. This skill defines track grouping, scene composition, caption placement, and safe-zone rules for Studio.
设计与创意 查看
📦
video-shots
76Analyze Videos Shot by Shot
由 eternityspring
Manual shot breakdowns are slow and timing estimates are unreliable. This skill measures cuts and motion, then produces structured reviews and interactive reports.
设计与创意 查看