Skill 詳細
travel-research
Useful for destination comparison before planning, but personalized to a specific traveler.
使用前に確認
自動レビューは関連性のみを確認し、安全性や推奨を保証しません。使用前に出典の説明を読んでください。
SKILL.md
これはレビュー時に保存された抜粋です。完全で最新の内容は外部ソースを確認してください。
--- name: travel-research description: Research and compare destinations with deeper scoring across budget, climate, logistics, activity fit, and seasonal tradeoffs before moving into full trip planning. version: 1.2.0 --- # Travel Research Skill ## Intent Use this skill when the user is not ready for a full itinerary yet and needs stronger destination research first. The goal is to compare places honestly, surface tradeoffs early, and identify the best fit before handing off into `travel-planning` or `travel-itinerary`. ## Use When - The user wants to compare destinations, regions, or trip styles - A trip idea is still in the research phase and the right destination is not settled - The user needs deeper analysis of climate, seasonality, training value, crowds, budget, or travel friction - The best next step is to narrow choices before building a day by day plan ## Do Not Use When - The destination is already chosen and the user needs a full itinerary - The request is mainly about booking specific flights, hotels, or transport - The user wants a finished travel budget and execution plan more than research - The question is a quick one off logistics check that does not need structured comparison ## Research Priorities Always judge options against Mick's actual travel preferences: - solo travel by default - private rooms only - BNA as the home airport unless otherwise stated - value optimized, not backpacker cheap and not luxury - preference for active trips, endurance anchors, and good photo worthy experiences Focus on what materially separates one option from another: - total trip cost and daily spend - weather and season fit - live or current flight pricing, weather normals, and crowd levels when available - event or activity alignment - transit complexity - travel friction score across visa effort, flight routing, and in-country transport difficulty - crowd pressure - safety and scam friction - visa or border hassle - how well the destination fits the stated trip goal ## Workflow 1. Identify the trip goal first. 2. Determine the comparison set: - user supplied destinations - a region shortlist - best fit suggestions if the user only gives goals and constraints 3. Build a direct scorecard for each option. 4. Pull live or current data when the request depends on timing-sensitive facts: - flight pricing from available public search results or user-provided fare data - weather normals from reliable climate or weather sources - crowd levels from seasonality, event calendars, booking pressure, or current travel sources 5. If live data is unavailable, label estimates clearly and explain what source or assumption was used. 6. Calculate a travel friction score for each destination from: - visa or entry effort - flight routing difficulty from BNA unless another origin is stated - in-country transport difficulty 7. Explain the tradeoffs in plain language, not tourism fluff. 8. Call out strong and weak fit areas: - budget fit - climate fit - activity fit - logistics fit - travel friction fit - pace fit - value for the stated season 9. Recommend the best option and at least one fallback. 10. State the next practical move: - deeper research - move into `travel-planning` - move into `travel-itinerary` ## Scoring Weight Presets Different trip types should not weight the scorecard the same way. Apply the preset that matches the trip goal and say which one was used: | Dimension | Endurance training | Sightseeing | Remote work | Budget recharge | |---|---:|---:|---:|---:| | Activity fit | 35% | 15% | 10% | 10% | | Climate | 25% | 15% | 10% | 20% | | Budget | 15% | 20% | 20% | 35% | | Logistics / friction | 15% | 25% | 25% | 20% | | Value for season | 10% | 25% | 35% | 15% | For mixed-goal trips, blend the two closest presets and note the blend. The weighted score feeds the Overall column; never let a destination win on vibe while losing on every weighted dimension. ##GitHub で全文を読む (外部ページ)