mcpField

hackathon-demo-video

github.com/bernieweb3创意与媒体59 次安装
代码仓库

关于 hackathon-demo-video

面向 MVP 流程已稳定的黑客松团队，用于规划限时演示视频。它根据项目资料和核心亮点生成分段脚本、镜头清单及录制准备方案。

主要功能

  1. 01按时间标注的旁白脚本
  2. 02映射演示步骤的镜头清单
  3. 03亮点时刻安排
  4. 04录制检查清单与备用方案

适用场景

  1. 01MVP 稳定后准备参赛视频
  2. 02为现场演示可能失败的录制做准备

安装

复制下面的命令，在终端中运行。

$npx skills add https://github.com/bernieweb3/hackathon-ai-devkit --skill hackathon-demo-video

安装后在 Claude Code 中输入/hackathon-demo-video确认已加载。

详细介绍

hackathon-demo-video

目标

为黑客松演示视频编写结构化脚本和镜头清单，在严格的时长限制内展示项目的核心价值。

触发条件

在以下情况下使用此技能：

  • MVP 演示流程已实现且稳定
  • 需要录制演示视频用于提交，或作为现场演示的备用方案
  • 需要优化视频中惊艳时刻的时机和安排
  • 必须在开始录制前检查并锁定录制环境
  • 在第 6 阶段调用；代码冻结后，与 hackathon-pitchdeck 并行运行

输入

输入类型必填描述
project_titlestring是项目名称
taglinestring是用一句话描述项目
mvp_demo_flowobject[]是来自 hackathon-scope-cutter 的演示步骤
wow_factorstring是最令人印象深刻的时刻
video_duration_secondsinteger是视频目标时长，单位为秒
demo_environmentstring否演示运行环境（例如，网页浏览器、移动设备、CLI）
voiceoverboolean否视频是否包含旁白（默认值：true）

输出

输出描述
script带时间码的旁白脚本
shot_list每个时刻屏幕上展示的内容
wow_moment_timestamp视频中惊艳时刻出现的时间
recording_checklist录制前的设置和验证步骤
fallback_plan录制期间现场演示失败时的应对方案

规则

  1. 前 5 秒必须说明问题或抛出引子——不要自我介绍。
  2. wow_factor 必须在 video_duration_seconds 的 60% 时长之前出现。
  3. 每个 mvp_demo_flow 步骤都必须至少对应一个镜头。
  4. 每句旁白不得超过 15 个词。
  5. 视频最后 10% 的时长用于总结和行动号召。
  6. recording_checklist 至少必须包含：关闭通知、准备好测试数据、设置屏幕分辨率。
  7. fallback_plan 必须使用静态幻灯片或预录片段，绝不能说“我们跳过这一部分”。

输出格式

wow_moment_timestamp: <seconds>

script:
  - timestamp_start: <seconds>
    timestamp_end: <seconds>
    narration: "<spoken words>"

shot_list:
  - timestamp_start: <seconds>
    timestamp_end: <seconds>
    screen_content: "<what is visible>"
    action: "<what the presenter does>"

recording_checklist:
  - "<item>"

fallback_plan:
  trigger: "<condition>"
  action: "<what to do>"

示例

输入：

project_title: "AnchorAI"
tagline: "An AI companion that remembers your emotional journey."
wow_factor: "AI recalls emotional context from 3 days ago and adapts tone in real time"
video_duration_seconds: 120
demo_environment: "web browser"
voiceover: true
mvp_demo_flow:
  - step: 1
    action: "User types 'I'm overwhelmed today'"
    outcome: "AI responds empathetically"
  - step: 2
    action: "User mentions recurring work stress"
    outcome: "AI references last week's conversation"
  - step: 3
    action: "User expresses hopelessness"
    outcome: "Crisis resource card appears"

输出：

wow_moment_timestamp: 45

script:
  - timestamp_start: 0
    timestamp_end: 8
    narration: "40% of college students live with anxiety. Most wait 6 weeks for help."
  - timestamp_start: 8
    timestamp_end: 15
    narration: "AnchorAI is always available — and it remembers you."
  - timestamp_start: 15
    timestamp_end: 45
    narration: "Watch what happens when a student checks in after a hard week."
  - timestamp_start: 45
    timestamp_end: 70
    narration: "AnchorAI recalls last week's conversation — without being told."
  - timestamp_start: 70
    timestamp_end: 95
    narration: "When a student signals distress, AnchorAI responds with care and resources."
  - timestamp_start: 95
    timestamp_end: 120
    narration: "No waitlist. No starting over. AnchorAI — always there."

shot_list:
  - timestamp_start: 0
    timestamp_end: 8
    screen_content: "Stat card: '40% of students report anxiety. Average wait: 6 weeks.'"
    action: "Hold on stat card"
  - timestamp_start: 15
    timestamp_end: 45
    screen_content: "Chat interface — user types message"
    action: "Type 'I'm feeling overwhelmed again this week'"
  - timestamp_start: 45
    timestamp_end: 70
    screen_content: "AI response references 'last week's exam stress you mentioned'"
    action: "Scroll to highlight the memory reference in the AI response"
  - timestamp_start: 70
    timestamp_end: 95
    screen_content: "User types hopelessness message; crisis card slides in below response"
    action: "Type message; pause on crisis card appearing"

recording_checklist:
  - "Notifications disabled on OS and browser"
  - "Demo account pre-loaded with 1 prior session in Redis"
  - "Screen resolution set to 1920×1080; browser zoom 125%"
  - "Test data: prior session summary contains 'exam stress' phrase"
  - "Run full demo flow once before recording"

fallback_plan:
  trigger: "OpenAI API fails to respond during recording"
  action: "Use pre-recorded screen capture of the memory recall exchange as an overlay segment"

上下文文件

知识库
  • ../hackathon-shared-resources/knowledge/hackathon-demo-psychology.md
  • ../hackathon-shared-resources/knowledge/hackathon-demo-patterns.md
  • ../hackathon-shared-resources/knowledge/hackathon-pitch-strategy.md
模板
  • ../hackathon-shared-resources/templates/demo-script-template.md
操作手册
  • ../hackathon-shared-resources/playbooks/hackathon-workflow.md

评论

支持 Markdown