← 返回 Skills 市场
yangagent

Whiteboard Video Workflow

作者 YangAgent · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
77
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install whiteboard-video-workflow
功能描述
从 SRT 字幕文件自动生成完整白板动画视频的端到端工作流。依次完成分镜解析、图片生成、视频生成三个阶段。当用户提供 SRT 文件并要求生成白板动画视频,或说"从字幕生成白板视频"、"白板视频工作流"时触发。
安全使用建议
This skill mostly does what it claims, but there are inconsistencies you should verify before installing or running it: - Metadata omission: The code expects a RUNNINGHUB_API_KEY (checked in check_env.py and used by generate-image.py) but the registry metadata does not declare any required environment variables. If you plan to use this skill you will need to provide a RunningHub API key (placed in <skill-dir>/.env or exported into environment). - Undeclared dependency on another skill: check_env.py tries to call a setup script in a sibling directory named 'whiteboard-animation' (scripts/setup_env.py). Confirm that this other skill and its setup_env.py exist and inspect that script before allowing it to run — it will install packages and execute arbitrary code. - Audit the setup/install scripts: check_env.py will attempt to auto-install dependencies via the referenced setup_env.py. Treat any install script as potentially dangerous until you read it. Run installs in a sandbox or container, or review the code and packages it installs. - Least-privilege for API key: if you provide a RunningHub key, prefer a key/account with minimal permissions and monitor usage. Consider testing with a dummy/limited key first. - Network activity: image generation contacts https://www.runninghub.cn and downloads result URLs. If you must avoid network exfiltration, do not run the scripts or run them in an isolated environment. - If you are not comfortable: request the publisher to update registry metadata to declare RUNNINGHUB_API_KEY and to document the dependency on whiteboard-animation (and ideally bundle or reference a vetted setup script). Otherwise classify this skill as moderate risk and only run after manual code review and in a controlled environment.
功能分析
Type: OpenClaw Skill Name: whiteboard-video-workflow Version: 0.1.0 The skill bundle provides a legitimate end-to-end workflow for generating whiteboard animation videos from SRT subtitle files. It coordinates multiple steps including storyboard parsing, AI image generation via the RunningHub API (scripts/generate-image.py), and video merging using the PyAV library (scripts/workflow_helper.py). While the skill performs network requests to 'www.runninghub.cn', downloads generated images, and manages Python dependencies via a setup script, these actions are transparently documented and strictly necessary for its stated functionality. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found.
能力评估
Purpose & Capability
The skill's purpose (SRT → whiteboard video) matches the included scripts (storyboard parsing, prompt generation, image/video generation). However the implementation requires a RunningHub API key and also calls a setup script located in a sibling directory 'whiteboard-animation' (check_env.py references ANIMATION_SKILL/scripts/setup_env.py). The package metadata declares no required env vars or skill dependencies — that's inconsistent and unexpected.
Instruction Scope
SKILL.md tightly defines steps and limits work to specific scripts/files, which is good, but instructions require: (a) running scripts that will auto-install dependencies (check_env.py will execute setup_env.py in another skill), (b) reading/writing absolute paths inside the skill tree, and (c) passing absolute paths/expanded values into subagents. The runtime will therefore execute local setup scripts and touch sibling-skill files — this expands scope beyond the single skill and is not declared.
Install Mechanism
There is no formal install spec (instruction-only), which is lower risk. But the runtime helper (check_env.py) will invoke a setup script (setup_env.py) to create/verify a Python virtualenv and install libraries; that installation happens at execution time and will run arbitrary code from the referenced setup script. The setup script lives outside this skill (whiteboard-animation), increasing uncertainty about what will be installed/executed.
Credentials
The scripts clearly require a RUNNINGHUB_API_KEY (generate-image.py posts to https://www.runninghub.cn and check_env.py insists on RUNNINGHUB_API_KEY in a .env file). Yet the skill metadata lists no required env vars or primary credential. Requesting and using an API key for an image-generation service is reasonable for this task, but the omission in metadata is a meaningful inconsistency. Also the skill expects access to a sibling skill's setup script, which is an undeclared external dependency.
Persistence & Privilege
The skill does not request 'always' or other elevated platform privileges. It runs subagents and local scripts but does not persistently modify other skills' configuration. The main concern is that check_env.py will execute setup_env.py (in another skill), which can modify the environment or install packages — review that script before running.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install whiteboard-video-workflow
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /whiteboard-video-workflow 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v0.1.0
Initial release – automate whiteboard video creation from SRT subtitles: - Fully automated end-to-end workflow: parses SRT to storyboard, generates images, and compiles video segments. - Robust environment check with clear user guidance for missing dependencies or API keys; auto-installs fixable items. - Modular step-based process; key phases (storyboard parsing, image generation, animation video creation) handled by subagents. - Strict sequence and file order validation to ensure accurate video output matching the SRT structure. - Unified intermediate data and result reporting in JSON for clarity and downstream use.
元数据
Slug whiteboard-video-workflow
版本 0.1.0
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 1
常见问题

Whiteboard Video Workflow 是什么?

从 SRT 字幕文件自动生成完整白板动画视频的端到端工作流。依次完成分镜解析、图片生成、视频生成三个阶段。当用户提供 SRT 文件并要求生成白板动画视频,或说"从字幕生成白板视频"、"白板视频工作流"时触发。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 77 次。

如何安装 Whiteboard Video Workflow?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install whiteboard-video-workflow」即可一键安装,无需额外配置。

Whiteboard Video Workflow 是免费的吗?

是的,Whiteboard Video Workflow 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Whiteboard Video Workflow 支持哪些平台?

Whiteboard Video Workflow 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Whiteboard Video Workflow?

由 YangAgent(@yangagent)开发并维护,当前版本 v0.1.0。

💬 留言讨论