Imported from amkessler/nicar2026_skills_in_codex_claude (.claude/skills/state-county-rankings/SKILL.md). Install upstream with npx skills add amkessler/nicar2026_skills_in_codex_claude --skill state-county-rankings. Copyright stays with the author.

State County Rankings

Use this skill to rank counties within a state for one or more metrics from a local dataset.

Requirements

• R with dplyr, tidyr, readr, stringr, and jsonlite
• Local CSV input with county, a state column (state, state_name, or state_abbrev), and requested metric columns
• Bundled Census CSV for immediate use: skills/state-county-rankings/data/county_demographics_acs5_2023.csv
• No API key required

Standard Workflow

1. Confirm input CSV path and state.
   • --state accepts full names or USPS abbreviations (for example, Georgia or GA).
   • Default bundled input: skills/state-county-rankings/data/county_demographics_acs5_2023.csv
2. Review skills/state-county-rankings/references/COLUMN_REQUIREMENTS.md and use exact output headers before downstream filtering/selecting.
   • Output columns are state, county, metric, value, rank, and direction.
3. Ask for metrics (or use defaults) and ranking direction.
4. Run:
   • Rscript skills/state-county-rankings/scripts/get_state_county_rankings.R --input skills/state-county-rankings/data/county_demographics_acs5_2023.csv --state "<state>"
5. Return top/bottom county rankings and note missing/invalid fields.

Output

Return ranked rows with:

• state
• county
• metric
• value
• rank
• direction