us-code

U.S. Code

The reader scripts automatically install the latest verified corpus on first use. After installation, lookups are local and the corpus is not checked for updates in that environment.

Use this skill when a task requires local U.S. Code text, section lookup, statutory citation checking, or statutory analysis based on federal code sections.

After first-use installation, lookup, listing, comparison, and search operate entirely against the local corpus.

Before the first command, locate the reader once. Cowork standalone skills normally use the first path below; the fallback also supports marketplace plugins and Claude Code:

SCRIPT="/root/.claude/skills/synced/us-code/scripts/read_section.py"
if [ ! -f "$SCRIPT" ]; then
  SCRIPT="scripts/read_section.py"
fi
if [ ! -f "$SCRIPT" ]; then
  SCRIPT="$(find /root/.claude /sessions -type f -path '*/us-code/scripts/read_section.py' -print -quit 2>/dev/null)"
fi
if [ -z "$SCRIPT" ] || [ ! -f "$SCRIPT" ]; then
  echo "Could not locate the installed U.S. Code reader." >&2
  exit 1
fi

Use python "$SCRIPT" ... for all commands. Do not run additional filesystem-wide searches unless this block fails.

Corpus

• references/index.json contains title metadata and corpus currency.
• references/titles/{titleKey}/index.json supports citation and heading lookup.
• Each title's sections.html contains its statutory section text.
• The first-use installer may normalize release files during extraction; the reader handles both canonical and distributed representations transparently.
• Appendix titles use keys such as 18-app.

Source checked August 6, 2026. The latest OLRC bulk preliminary release available on that date was current through Public Law 119-102, July 12, 2026. An installed corpus does not update itself at runtime.

This corpus includes statutory section text and section metadata only. Source credits, amendment and effective-date notes, transfer and historical notes, and other editorial material are not included.