suzheng/k2f

k2f-travel-planner

Blueprint a printable, trip-ready travel planner from destination, dates, budget, and must-dos—day-by-day flow, budget roll-up, packing, culture notes, and prep timeline with scannable hierarchy.

ソースを見る

使い方

  1. インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
  2. 必要なツール、認証情報、対応環境を原文で確認します。
  3. 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/suzheng/k2f --skill k2f-travel-planner

開始前の確認

  • 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
  • スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
リポジトリの原文

見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。

K2F Travel Planner — AI-Native Aesthetic Design

Purpose

Teach agents to author travel planners in content_and_design.md: trip content, information layering, and design intent so the result reads like a field guide—not a chat log or destination encyclopedia.

Built on K2F

This skill produces travel planners as .K2F packages. Follow the k2f skill for authoring, pack/verify, and export. K2F keeps agents on trip semantics and visual rhythm while the engine handles layout and typography.

Before you build, install the core skill: npx skills add suzheng/k2f --skill k2f.

When to Use

  • Turning trip constraints into a shareable planner (couple, family, or group) with days, costs, and checklists in one document.
  • Merging user notes, bookings, and research into one coherent structure.
  • Not for destination-only chat or live expense tracking unless the deliverable is still a planner blueprint.

Workflow

Step 1. Generate the contentanddesign.md

One blueprint: semantic blocks per planner section plus explicit design intent. Label sections so the k2f skill can map them later.

Sub-steps

  1. Trip frame — Destination(s), dates, duration, travelers, purpose, budget (total or daily), currency, pace, accommodation base(s), must-sees. Note routing (linear, loop, hub-and-spoke) and transit or jet-lag recovery days.
  1. Input audit — List confirmed bookings, dietary or accessibility needs, interests to weight, and gaps. Mark NEEDS USER INPUT for unknown flight times, hotel address, or ticket status; do not invent opening hours, prices, or visa rules—use verify before travel stubs.
  1. Section architecture (include / cut / merge) — Default stack (reorder when the frame demands):
  • Trip at a glance (dates, party, budget headline, base neighborhoods) → Essentials (documents, emergency, money, connectivity, key transport) → Day-by-day itinerary → Budget summary → Packing checklist → Culture & safety (compact) → Pre-trip countdown.
  • Cut long destination essays, duplicate restaurant lists, generic packing unrelated to climate/activities, and hour-by-hour filler when pace is relaxed.
  • Merge overlapping transport tips into Essentials; split multi-city trips into labeled day ranges with transit days called out explicitly.
  1. Itinerary build — Theme per day; geo-cluster stops; time blocks with duration, transit, meals, cost, booking notes. Pacing: short trips 2–3 majors/day; week-long 2 major + 1 minor plus one light day; buffers and an optional rain/tired swap every 2–3 active days.
  1. Budget reconciliation — Fixed costs (flights, lodging, insurance, visas) plus daily buckets (food, local transport, activities, misc); 10% contingency; show total, per-person, and per-day; line items must trace to itinerary activities or stated assumptions.
  1. Supporting modules — Packing grouped Documents / Clothing (climate + dress codes) / Activity gear; culture as short do/don’t and dining timing; prep timeline working backward from departure (bookings that sell out first at the top).
  1. Scan layers — Glance: where, when, how much headroom. Scan: day titles and time leaders. Detail: addresses, confirmation hints, costs—subordinate in design intent.
  1. Design intent — Tone: practical, warm, local-aware; imperative checklists for actions. Hierarchy: title and dates dominate; day + theme anchors; times and places beat prose; costs and must-book flags scannable. Density: airy overview; consistent day blocks; checkbox lists; culture as callouts. Palette: calm journal feel; one accent on day headers; strong contrast on times and bookings.

Rules & constraints

  • Truth gate: prices, hours, and requirements trace to user input or labeled stubs—not guessed from memory.
  • Pace gate: total planned hours fit the stated pace; include meals and transit in every day.
  • Budget gate: category totals reconcile with headline budget or show explicit overrun note.
  • Routing gate: minimize backtracking; high-energy days not stacked after red-eye arrival without recovery.
  • Planner boundary: if the user only wants a spreadsheet tracker, note in blueprint that this skill targets a designed planner artifact.

Common failure modes

  • Attraction list without time or geography → themed days with clustered blocks.
  • Packed schedule with no rest → pacing table and one light day per week-long trip.
  • Budget section disconnected from days → tie line items to named activities.
  • Packing encyclopedia → climate + activity-filtered checklist only.
  • Culture section as generic country essay → 5–8 actionable do/don’t plus dining timing.
  • Prep timeline after departure or missing sell-out bookings → reverse chronological from departure date.
  • Repeating the same practical tip in every day → lift once to Essentials.

Quality checklist

  • [ ] Frame: destination, dates, party, budget, pace, routing, and must-sees documented.
  • [ ] Section stack fits trip length; cuts and multi-city splits explained.
  • [ ] Each day has theme, geo-logical flow, meals, transport, costs, and booking notes or stubs.
  • [ ] Budget reconciles (or flags gap) with contingency stated.
  • [ ] Packing and prep match climate, activities, and departure date.
  • [ ] Glance layer answers where/when/how much without reading full days.
  • [ ] Design intent covers tone, hierarchy, density, checklist rhythm, and accent.

Step 2. Use the contentanddesign.md and k2f skill to generate a K2F file and export to an appropriate file format.

Follow k2f to instantiate planner sections from the blueprint, apply roles from design intent, verify, and export (typically PDF or print-ready format).

タスクに関連する項目

同じリポジトリから

関連する Skills

すべての Skills
suzheng
コミュニティ

k2f-resume

Turn career history, notes, or an existing CV into a tailored, achievement-led resume with clear hierarchy and scannable proof—not a generic duty list or keyword-stuffed template. Use when the user asks for a resume, CV, job application document, or role-targeted career summary (including refresh for a specific job posting). Create resumes with AI-native design and the highest aesthetic standards, powered by the K2F format and engine. Purpose-built for AI agents, K2F lets them focus on semantic

導入数
10
GitHub Stars
3
更新日
10月15日
suzheng
コミュニティ

k2f-worksheet

Design printable, fill-in-ready worksheets—practice sets, graphic organizers, lab records, reflections, or gap-closure forms—from learning goals and source material. Use when the user asks for a worksheet, handout with answer spaces, practice problems, graphic organizer, or structured questions students complete in class or at home. Create worksheets with AI-native design and the highest aesthetic standards, powered by the K2F format and engine. Purpose-built for AI agents, K2F lets them focus o

導入数
10
GitHub Stars
3
更新日
10月15日
suzheng
コミュニティ

k2f-grant-proposal

Turn research ideas, preliminary data, and agency solicitations into reviewer-ready grant proposals with a fundable hypothesis, independent specific aims, and agency-aligned significance and innovation. Use when the user asks for a grant proposal, specific aims page, NIH R01/R21/K, NSF proposal, or foundation grant narrative. Create grant proposals with AI-native design and the highest aesthetic standards, powered by the K2F format and engine. Purpose-built for AI agents, K2F lets them focus on

導入数
10
GitHub Stars
3
更新日
10月15日
suzheng
コミュニティ

k2f-media-kit

Turn verified company, product, and leadership inputs into a journalist-ready media kit with scannable facts, boilerplate, and asset guidance—not a pitch deck or sales one-pager. Use when the user asks for a media kit, press kit, PR kit, or creator resource pack. Create media kits with AI-native design and the highest aesthetic standards, powered by the K2F format and engine. Purpose-built for AI agents, K2F lets them focus on semantic content and design intent while automatically handling layou

導入数
8
GitHub Stars
3
更新日
10月15日