hackathon-task-planner

hackathon-task-planner

目标

将 MVP 范围拆解为有序且有时间限制的任务清单，分配角色并明确依赖关系，以便在黑客松期间执行。

---

触发条件

在以下情况下使用此技能：

• 可从 hackathon-scope-cutter 获取带时间预算的 MVP 功能
• 团队需要在开始编码前制定结构化的任务清单
• 需要定义角色和并行工作流
• 必须确定关键路径，以保障演示时间安排
• 在第 4 阶段为每个项目调用一次；如果实施期间缩减范围，则重新调用

---

输入

输入	类型	必填	描述
mvp_features	object[]	是	来自 hackathon-scope-cutter、带有时间预算的 MVP 功能
tech_stack	string[]	是	使用的技术
team_size	integer	是	团队成员人数
team_roles	string[]	否	角色标签（例如，前端、后端、ML、设计）
hackathon_duration_hours	integer	是	可用的总时长
start_offset_hours	integer	否	黑客松开始后已过去的小时数（默认值：0）

---

输出

输出	描述
tasks	包含估算时间、角色和依赖关系的完整任务清单
critical_path	决定项目完成进度的有序任务序列
milestones	带有目标小时数的关键检查点
parallel_tracks	可同时进行的任务组
buffer_hours	为集成、润色和调试预留的小时数
recommended_skills	建议接下来调用的技能

---

规则

1. 将每项 MVP 功能拆解为耗时 30 分钟至 3 小时的任务。
2. 明确列出搭建、集成和部署任务。
3. 将每项任务恰好分配给 team_roles 中的一个角色（若未指定，则分配给“any”）。
4. 将关键路径确定为最长的依赖链。
5. 预留的 buffer_hours 至少为黑客松剩余时间的 15%。
6. 将 milestones 依次安排在剩余时间的 25%、50%、75% 和 90% 处。
7. 将所有没有明确负责人的任务标记为 [UNASSIGNED]。

---

输出格式

tasks:
  - id: "T-<number>"
    title: "<task title>"
    feature: "<parent feature>"
    role: "<role|[UNASSIGNED]>"
    estimated_hours: <number>
    depends_on:
      - "T-<number>"

critical_path:
  - "T-<number>"

milestones:
  - name: "<milestone name>"
    target_hour: <number>
    deliverable: "<what must exist>"

parallel_tracks:
  - track: "<track name>"
    tasks:
      - "T-<number>"

buffer_hours: <number>

recommended_skills:
  - "<skill-name>"

---

示例

输入：

mvp_features:
  - feature: "GPT-4 check-in conversation"
    estimated_hours: 4
  - feature: "Session memory (Redis)"
    estimated_hours: 5
  - feature: "Crisis escalation (mocked)"
    estimated_hours: 1
tech_stack: ["Python", "FastAPI", "React", "OpenAI API", "Redis"]
team_size: 3
team_roles: ["backend", "frontend", "fullstack"]
hackathon_duration_hours: 24
start_offset_hours: 2

输出：

tasks:
  - id: "T-01"
    title: "Set up FastAPI project and Redis"
    feature: "Session memory"
    role: "backend"
    estimated_hours: 1
    depends_on: []
  - id: "T-02"
    title: "Implement OpenAI chat endpoint"
    feature: "GPT-4 check-in conversation"
    role: "backend"
    estimated_hours: 2
    depends_on: ["T-01"]
  - id: "T-03"
    title: "Implement session memory read/write"
    feature: "Session memory"
    role: "backend"
    estimated_hours: 3
    depends_on: ["T-01"]
  - id: "T-04"
    title: "Build chat UI component"
    feature: "GPT-4 check-in conversation"
    role: "frontend"
    estimated_hours: 3
    depends_on: []
  - id: "T-05"
    title: "Wire frontend to chat API"
    feature: "GPT-4 check-in conversation"
    role: "fullstack"
    estimated_hours: 1
    depends_on: ["T-02", "T-04"]
  - id: "T-06"
    title: "Add crisis card component (mocked)"
    feature: "Crisis escalation"
    role: "frontend"
    estimated_hours: 1
    depends_on: ["T-04"]

critical_path:
  - "T-01"
  - "T-02"
  - "T-05"

milestones:
  - name: "Backend skeleton live"
    target_hour: 7
    deliverable: "FastAPI + Redis running; chat endpoint returns GPT-4 response"
  - name: "Frontend connected"
    target_hour: 11
    deliverable: "Chat UI communicates with backend end-to-end"
  - name: "Memory working"
    target_hour: 16
    deliverable: "AI recalls previous session context in demo flow"
  - name: "Demo-ready"
    target_hour: 19
    deliverable: "Full demo flow runs cleanly 3× in a row"

parallel_tracks:
  - track: "Backend"
    tasks: ["T-01", "T-02", "T-03"]
  - track: "Frontend"
    tasks: ["T-04", "T-06"]

buffer_hours: 3.3

recommended_skills:
  - "hackathon-code-implementer"

---

上下文文件

知识库

• ../hackathon-shared-resources/knowledge/hackathon-mvp-strategy.md
• ../hackathon-shared-resources/knowledge/hackathon-reference-architecture.md
• ../hackathon-shared-resources/knowledge/hackathon-common-failures.md
• ../hackathon-shared-resources/knowledge/hackathon-tools.md

操作手册

• ../hackathon-shared-resources/playbooks/hackathon-workflow.md
• ../hackathon-shared-resources/playbooks/24h-hackathon-playbook.md
• ../hackathon-shared-resources/playbooks/36h-hackathon-playbook.md
• ../hackathon-shared-resources/playbooks/48h-hackathon-playbook.md