plugins/r-econ-data/skills/validate-analysis/SKILL.md
Common ways to validate or benchmark data analysis estimates using external sources. Use to confirm or test the plausibility of results from an economic analysis.
npx skillsauth add Economic/epi-skills validate-analysisInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
For example, say you wanted to calculate the real median hourly wage for a demographic group not available in SWADL, like foreign-born workers in California. To verify your basic approach, you might first try to replicate something in SWADL approximating the target group, like the real median wages of all workers in California:
library(tidyverse)
library(epidatatools)
library(epiextractr)
library(realtalk)
library(swadlr)
library(assertr)
swadl_ca_median = get_swadl(
"hourly_wage_median",
measure = "real_wage_median_2025",
geography = "CA"
) |>
mutate(year = year(date)) |>
select(year, swadl_real_median = value)
cpi_data = c_cpi_u_annual
cpi_base = cpi_data |>
filter(year == 2025) |>
pull(c_cpi_u)
org_median_wage = load_org(
2010:2025,
year,
statefips,
wage,
orgwgt,
citistat
) |>
filter(wage > 0, statefips == 6) |>
summarize(
org_nominal_median = averaged_median(wage, w = orgwgt),
.by = year
) |>
left_join(cpi_data, by = "year") |>
mutate(org_real_median = org_nominal_median * cpi_base / c_cpi_u)
org_median_wage |>
left_join(swadl_ca_median, by = "year") |>
verify(swadl_real_median - org_real_median < 1e-8)
tools
--- name: research-remote-code description: Techniques for retrieving and investigating remote codebases and their documentation. Use when user provides a GitHub/GitLab URL to look at, asks to follow conventions from a remote repo, asks "how does package/library X implement/handle Y", or references any external repository's code, structure, or patterns. Examples: "follow the conventions in https://github.com/...", "look at how repo X does Y", "check this GitHub link". --- ## Accessing code from
development
Provides useful economic data sources and ways to retrieve the data. Use when user asks for "prices", "inflation", "employment", "unemployment", "inequality", "productivity", "poverty", "CPS", "Census", "BLS", "BEA", "FRED", "IPUMS", "SWADL", "State of Working America Data Library".
testing
Methods for calculating statistics and estimates using economic data. Use when user asks to calculate or estimate quantities like percentiles, means, medians, shares, counts, or regression coefficients or regression-adjusted statistics.
testing
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".