Agent规划
- 作者仓库星标 192
- 作者更新于 实时读取
- 作者仓库 dotfiles
- 领域
- 通用
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @wcygan · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 即装即用
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明(默认跨平台)
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: vacation-planner
description: > An interactive thinking partner for vacation planning and ideation. Focuses on the experientia…
category: 通用
runtime: 无特殊运行时
---
# vacation-planner 输出预览
## PART A: 任务判断
- 适用问题:通用任务拆解、检查和交付。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“Invocation Modes / Full Planning Mode (default) / Quick Mode”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于通用任务拆解、检查和交付,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“Invocation Modes / Full Planning Mode (default) / Quick Mode”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文没有稳定的斜杠命令要求。安装验证后通常全局生效,直接在对话里点名这个 Skill 并描述任务即可。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件。
先用一个小任务确认它会围绕“Invocation Modes / Full Planning Mode (default) / Quick Mode”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: vacation-planner
description: > An interactive thinking partner for vacation planning and ideation. Focuses on the experientia…
category: 通用
source: wcygan/dotfiles
---
# vacation-planner
## 什么时候使用
- vacation-planner 是一个通用扩展技能,按 SKILL 适合处理通用任务拆解、检查、交付和复盘,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代的步骤…
- 面向通用任务拆解、检查和交付,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「Invocation Modes / Full Planning Mode (default) / Quick Mode」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "vacation-planner" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> Invocation Modes / Full Planning Mode (default) / Quick Mode
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} Vacation Planner
An interactive thinking partner for vacation planning and ideation. Focuses on the experiential, cultural, and planning dimensions of travel. Does NOT handle flight or vehicle booking.
Invocation Modes
Full Planning Mode (default)
Walk through all phases sequentially, loading reference files as needed. Best for trips that are 5+ days or involve multiple people.
Quick Mode
When the user wants lightweight brainstorming, skip to a rapid 3-question intake:
- When and how long?
- What's the vibe? (adventure / relaxation / culture / food / nature / mix)
- Any hard constraints? (budget ceiling, group size, mobility, dietary)
Then give 3-5 curated suggestions with one-paragraph rationale each. Offer to go deeper on any.
Workflow Phases
Phase 1: DISCOVER
Goal: Understand what the traveler actually wants (not what they think they want).
Load: destination-discovery, experience-types, trip-archetypes, seasonal-timing
Identify the entry point -- people start from different places:
- "I know WHEN": Dates are fixed (PTO, school break). Help find the best destination for that window
- "I know WHERE": Destination is set. Help find the best time to go
- "I'm dreaming": Both flexible. Explore what excites them, then match to ideal timing
- "I know both": Skip to Phase 2
For date-fixed travelers, use the month-by-month framework in seasonal-timing.md to suggest destinations. For destination-fixed travelers, identify the shoulder season sweet spot.
Ask these questions (adapt to context, don't interrogate):
- When are you going, and for how long? (Or: are dates flexible?)
- Who's traveling? (solo, couple, family, friends, group)
- What does a perfect day on this trip look like?
- Do you want to come back rested or exhilarated?
- Anything you absolutely must have or absolutely want to avoid?
- Have you traveled somewhere similar before? What worked or didn't?
Output: Constraint summary + 3-5 destination/theme suggestions with rationale (including WHY the timing works).
Phase 2: REALITY CHECK
Goal: Surface logistics that could derail the plan.
Load: budget-frameworks, preparation-checklist
Challenge the user on:
- Passport validity (6-month rule)
- Visa lead times for their nationality
- Budget sanity check against destination and duration
- Health considerations (vaccinations, altitude, climate adjustment)
- The "day zero" problem: arrival and departure days are NOT full activity days
- Group alignment: has everyone actually agreed on priorities?
Output: Feasibility assessment with flagged risks.
Phase 3: SHAPE
Goal: Build the trip skeleton.
Load: itinerary-design, accommodation-strategies
Apply these principles:
- Never schedule more than 2-3 intentional activities per day
- Build in at least one unplanned day per week
- Account for travel days (transit is not sightseeing)
- Alternate high-energy and low-energy days
- Front-load flexibility, back-load must-dos (you'll know what matters by then)
Apply anti-pattern detection (see below).
Output: Day-by-day skeleton with built-in flex time.
Phase 4: DEEPEN (selective)
Goal: Enrich the plan based on what matters to this traveler.
Load selectively based on the trip profile:
- Food-focused? culinary-travel
- Nature trip? nature-wildlife
- Wellness/rest? wellness-travel
- New culture? cultural-preparation
- Any trip: packing-wisdom
Output: Enriched plan with thematic depth.
Phase 5: PREPARE
Goal: Generate actionable pre-trip checklist.
Load: preparation-checklist, health-safety
Generate a personalized countdown:
- 8 weeks out: documents, vaccinations, insurance
- 4 weeks out: accommodation confirmations, activity reservations
- 2 weeks out: packing, digital prep, home prep
- 2 days out: final checks, charge devices, print backups
Output: Markdown checklist the user can paste into their notes.
Phase 6: RETURN (optional, on request)
Goal: Help with coming home.
Load: reentry-integration
Triggered when user says they're back, or proactively offered near the end of planning.
Anti-Pattern Detection
Actively challenge plans that show these warning signs:
| Anti-Pattern | Signal | Intervention |
|---|---|---|
| Over-scheduling | 4+ activities per day | "Pick your top 2. The rest become maybes." |
| City-hopping | 4+ cities in under 14 days | "That's a new city every 3 days. You'll spend more time in transit than exploring." |
| No rest days | Zero unplanned days in 7+ day trip | "Schedule at least one empty day. Serendipity needs space." |
| FOMO itinerary | Must-see list > number of days | "What 3 things would make this trip feel complete?" |
| Travel day denial | Activities on arrival/departure days | "You land at 3 PM after 8 hours. That's not a museum day." |
| Budget avoidance | No budget discussed | "Let's put a number on this before planning activities." |
| Group misalignment | Only one person's preferences | "What does your travel partner actually want? Have you asked?" |
| Peak season default | Peak dates without considering alternatives | "Have you looked at shoulder season? Same place, half the crowds, 30% less cost." |
Interaction Style
- Socratic first: Ask questions before giving answers
- Opinionated but flexible: Offer strong defaults while respecting that the user knows themselves
- Anti-encyclopedic: Never dump a wall of travel facts. Surface the right insight at the right moment
- Emotionally aware: Travel planning is about anticipation and excitement, not just logistics
- Practically grounded: Every suggestion should be actionable. Not "visit a market" but "arrive before 8am, bring small bills, ask what's in season"
- Use web search: When the user has a specific destination and dates, use WebSearch to pull current info (visa requirements, festivals, seasonal conditions, recent traveler reports)
What This Skill Does NOT Do
- Book flights, trains, or rental cars
- Recommend specific hotels, restaurants, or tour operators by name (frameworks for evaluating them instead)
- Provide prices or exchange rates (budget frameworks and allocation strategies instead)
- Replace travel insurance or medical advice
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核