Run2 travel data extraction

Skill cxcscmu/SkillLearnBench/skills/b2-self-feedback-claude-opus-4-6/travel-planning/run2_travel-data-extraction

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill run2_travel-data-extraction

Assembled 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

Extract and validate travel data from CSV/TXT databases ensuring city-specific matching and constraint compliance.

SKILL.md

1.6 KB, as published. Nobody here has run it

Travel Data Extraction (Improved)

Data Sources

data/background/citySet_with_states.txt - Tab-separated city/state pairs
data/accommodations/clean_accommodations_2022.csv - Columns: index, NAME, room type, price, minimum nights, review rate number, house_rules, maximum occupancy, city
data/restaurants/clean_restaurant_2022.csv - Columns: index, Name, City, Cuisines, Average Cost, Aggregate Rating
data/attractions/attractions.csv - Columns: Name, Latitude, Longitude, Address, Phone, Website, City
data/googleDistanceMatrix/distance.csv - Columns: origin, destination, cost, duration, distance

Critical: City-Specific Validation

When searching for entities, ALWAYS verify the city column matches. Names can appear in multiple cities:

# WRONG - may match wrong city
grep "Restaurant Name" restaurants.csv
# RIGHT - verify city match
grep "Restaurant Name.*TargetCity" restaurants.csv

Pet-Friendly Accommodations

Filter: rows where house_rules does NOT contain "No pets". Also check:

maximum_occupancy >= party size (2 for couples)
minimum_nights <= planned stay length
Price fits within budget

Cuisine Matching

The Cuisines field is comma-separated. Match keywords: American, Mediterranean, Chinese, Italian. Ensure all 4 cuisine types appear across the itinerary's meals.

Distance Data

Self-driving cost column is always empty
Duration format: "X hours Y mins" or "X mins"
Distance format: "X,XXX km" with comma separators