Itinerary budget planning
Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-opus-4-6/travel-planning/itinerary-budget-planning
[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
npx -y skills add cxcscmu/SkillLearnBench --skill itinerary-budget-planningAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its author says it does
Copied from the file, not written here
Planning multi-city travel itineraries with budget constraints, route optimization, and cuisine diversity.
SKILL.md
1.5 KB, 294 tokens by cl100k_base, as published. Nobody here has run it
Itinerary Budget Planning
Route Optimization
- Choose cities that minimize total driving time
- Consider geographic clustering (e.g., Cleveland→Dayton→Cincinnati forms a logical loop)
- Start and end at the origin city for round trips
Budget Components
- Accommodations: price × nights per city
- Meals: restaurant average cost × number of meals
- Transportation: Self-driving costs (gas estimated from distances)
Cuisine Diversity Strategy
- Map required cuisines (American, Mediterranean, Chinese, Italian) across meals
- Use restaurant
Cuisinesfield to match — many restaurants serve multiple cuisines - Spread different cuisines across breakfast, lunch, and dinner slots
Accommodation Selection Criteria
- Check
maximum occupancy >= party size - Check
minimum nights <= planned stay duration - Filter by
house_rulesfor specific needs (pets, children, smoking) - Balance price vs. review rating
Day Planning Template
- Travel days: skip meals during long drives (use "-")
- City days: 3 meals + 2-3 attractions
- Transition days: breakfast in departing city, lunch/dinner in arriving city
Output Format
JSON with plan array (7 day objects) and data_sources array.
Each day: day, current_city, transportation, breakfast, lunch, dinner, attraction, accommodation.