mcpField

hackathon-demo-script

github.com/bernieweb3营销与增长43 次安装
代码仓库

关于 hackathon-demo-script

hackathon-demo-script 是用于准备黑客松现场演示的脚本技能，面向需要排练展示流程的团队。它生成 60–90 秒叙述、逐步操作与讲解，并安排主要亮点出现的时机。

主要功能

  1. 01生成 60–90 秒演示叙述
  2. 02逐步编排操作和口播
  3. 03标注步骤时长与转场提示
  4. 04安排主要亮点的呈现时机
  5. 05生成排练检查清单

适用场景

  1. 01为评审准备现场演示
  2. 02让多位演讲者分配讲解
  3. 03排练稳定一致的演示流程

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-script

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

详细介绍

hackathon-demo-script

目标

为黑客松演示生成一段可排练的 60–90 秒演示叙述，其中包括逐步演示流程、每个步骤的口述台词，以及主要惊艳时刻的明确安排，以最大程度打动评委。

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
  • 需要现场演示讲稿（与录制视频讲稿不同）
  • 演示者需要口述台词和现场演示的过渡提示
  • 团队必须在评审前排练一致的叙述
  • 在第 6 阶段（演示准备）调用，与 hackathon-pitchdeck 并行；配合 hackathon-demo-video 用于现场演示

输入

输入类型必填描述
project_titlestring是项目名称
taglinestring是一句话项目介绍
problem_statementstring是要解决的问题
mvp_demo_flowobject[]是来自 hackathon-scope-cutter 的演示步骤
primary_wow_momentstring是来自 hackathon-wow-detector、最令人印象深刻的时刻
target_userstring是主要用户群体
evaluation_axesobject[]是来自 hackathon-track-analyzer 的评审标准
demo_duration_secondsinteger否目标演示时长（秒，默认值：75）
presenter_countinteger否演示者人数（默认值：1）；用于拆分讲述内容

输出

输出描述
demo_narrative从开场到结束的完整演示叙述
demo_flow逐步演示内容，包括口述台词、屏幕操作、时间安排和每页时长
wow_moment_staging主要惊艳时刻的确切安排说明
transition_cues在各演示步骤之间自然过渡的口头提示
rehearsal_checklist现场演示前需要确认的事项
recommended_skills建议接下来调用的技能

规则

  1. demo_flow 的总时长必须在 demo_duration_seconds ± 10 秒范围内。
  2. 开场要从问题或用户故事切入，绝不能说“嗨，我们是 X 团队，我们开发了……”
  3. primary_wow_moment 必须出现在 demo_duration_seconds 的 60% 之前。
  4. mvp_demo_flow 中的每个步骤都必须对应至少一个 demo_flow 步骤。
  5. 每个步骤的口述台词不得超过 15 个词，以便在操作界面时也能自信地简短讲述。
  6. demo_flow 的每个步骤都必须包含 estimated_duration_seconds，帮助演示者控制节奏。
  7. wow_moment_staging 必须包括：暂停指示、要说的话、观众看到的内容。
  8. rehearsal_checklist 必须包括：演示账号状态、测试数据已加载、浏览器/应用状态、通知已关闭。

输出格式

demo_narrative: |
  <Full prose narrative — 3–5 sentences summarising the complete demo arc>

demo_flow:
  - step: <number>
    timestamp_start_seconds: <number>
    timestamp_end_seconds: <number>
    estimated_duration_seconds: <number>
    screen_action: "<what to do on screen>"
    spoken_lines: "<what to say>"
    presenter: "<presenter 1|presenter 2|all>"
    is_wow_moment: <true|false>

wow_moment_staging:
  step: <number>
  setup: "<what to say or do immediately before>"
  action: "<the exact on-screen action>"
  pause_seconds: <number>
  what_audience_sees: "<description>"
  spoken_reveal: "<what to say after the pause>"

transition_cues:
  - from_step: <number>
    to_step: <number>
    spoken_cue: "<bridging phrase>"

rehearsal_checklist:
  - "<item>"

recommended_skills:
  - "<skill-name>"

示例

输入：

project_title: "AnchorAI"
tagline: "An AI companion that remembers your emotional journey."
problem_statement: "College students wait 6 weeks for counseling while anxiety compounds daily."
mvp_demo_flow:
  - step: 1
    action: "User types 'I'm feeling overwhelmed today'"
    outcome: "AI responds empathetically"
  - step: 2
    action: "User describes recurring work stress"
    outcome: "AI references last week's conversation from memory"
  - step: 3
    action: "User expresses hopelessness"
    outcome: "Crisis resource card appears"
primary_wow_moment: "AI recalls emotional context from 3 days ago and adapts tone in real time"
target_user: "College students with anxiety"
evaluation_axes:
  - axis: "Innovation"
    weight: "high"
  - axis: "Impact"
    weight: "high"
demo_duration_seconds: 75
presenter_count: 1

输出：

demo_narrative: |
  AnchorAI solves a simple problem: 40% of college students live with anxiety, but the average campus counseling wait is 6 weeks. AnchorAI provides always-available, context-aware emotional support — and unlike every other AI tool, it actually remembers you. Watch a student check in after a hard week, and see how AnchorAI picks up exactly where they left off.

demo_flow:
  - step: 1
    timestamp_start_seconds: 0
    timestamp_end_seconds: 12
    estimated_duration_seconds: 12
    screen_action: "Open AnchorAI chat interface; type 'I'm feeling overwhelmed today'"
    spoken_lines: "Meet Alex. It's exam week. Alex opens AnchorAI and checks in."
    presenter: "presenter 1"
    is_wow_moment: false

  - step: 2
    timestamp_start_seconds: 12
    timestamp_end_seconds: 42
    estimated_duration_seconds: 30
    screen_action: "Type 'I keep worrying about the same things as last week'"
    spoken_lines: "Alex mentions the same stress from last week. Watch what AnchorAI does."
    presenter: "presenter 1"
    is_wow_moment: true

  - step: 3
    timestamp_start_seconds: 42
    timestamp_end_seconds: 65
    estimated_duration_seconds: 23
    screen_action: "Type 'I feel completely hopeless'; observe crisis card appear"
    spoken_lines: "When Alex signals distress, AnchorAI responds with care — and resources."
    presenter: "presenter 1"
    is_wow_moment: false

  - step: 4
    timestamp_start_seconds: 65
    timestamp_end_seconds: 75
    estimated_duration_seconds: 10
    screen_action: "Hold on the full screen"
    spoken_lines: "No waitlist. No starting over. AnchorAI — always there."
    presenter: "presenter 1"
    is_wow_moment: false

wow_moment_staging:
  step: 2
  setup: "Pause after typing the message. Don't speak for 2 seconds while the response loads."
  action: "Scroll slowly to highlight the sentence where AnchorAI references 'last week's exam stress you mentioned'"
  pause_seconds: 3
  what_audience_sees: "AI response contains a specific reference to prior session content — without being prompted"
  spoken_reveal: "AnchorAI remembered. Without being told. That's the difference."

transition_cues:
  - from_step: 1
    to_step: 2
    spoken_cue: "Alex has been here before. Let's see if AnchorAI remembers."
  - from_step: 2
    to_step: 3
    spoken_cue: "But what happens when things get harder?"
  - from_step: 3
    to_step: 4
    spoken_cue: "That's AnchorAI. Built for the moments between appointments."

rehearsal_checklist:
  - "Browser notifications disabled (OS and Chrome)"
  - "Demo account pre-loaded with 1 prior Supabase session containing 'exam stress' phrase"
  - "Chat interface loaded and visible at correct zoom level (125%)"
  - "Groq API key verified active with a test request"
  - "Run full demo flow twice before judging; verify wow moment appears consistently"
  - "Fallback screen recording ready in case of API failure"

recommended_skills:
  - "hackathon-pitchdeck"
  - "hackathon-judge-simulator"

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
  • ../hackathon-shared-resources/knowledge/hackathon-winning-patterns.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