Skip to content

vale-cli/agent-tools / skills/vocab

Add project terms to a Vale vocabulary so spell check accepts them, and set up a regional dictionary when the team wants British, Australian, Canadian or American spelling enforced. Use when the user says "vale doesn't know our product name", "stop flagging this word", "we write in British English", "flag American spellings", or spelling alerts are firing on correct terms.

FileKindSize
SKILL.mdskill md8.2 KB
scripts/dictionary.shexecutablescript6.1 KB
scripts/import-words.shexecutablescript4.6 KB

scripts/dictionary.sh

View on GitHub
#!/bin/sh
#
# Work out which English a project writes in, and install the dictionary
# that enforces it.
#
#   dictionary.sh -d path...             count American and British spellings
#   dictionary.sh [-c .vale.ini] en_GB   install a regional English dictionary
#   dictionary.sh [-c .vale.ini] fr      install another language's dictionary
#
# Vale's built-in dictionary accepts both "color" and "colour", so a project
# gets no spelling noise from either -- and no help keeping to one. A regional
# dictionary makes the other variant a misspelling. It is also a general
# dictionary: it does not know "plugin" or "config" the way the built-in one
# does, so expect the vocabulary to grow when you switch.
#
# For a language other than English the dictionary is what makes spell check
# work at all: Vale carries none, and skips files in a language it has no
# dictionary for. Name the language as the config's `Lang` key does -- fr, de,
# es, pt-BR -- and the files are saved under the name Vale looks for.
#
# The dictionaries are Hunspell files from the `wooorm/dictionaries`
# collection (about ninety languages), downloaded into
# <StylesPath>/config/dictionaries, which is source and belongs in the
# repository. Their licences vary by language; each one's is saved beside it.
#
# POSIX sh and POSIX tools only, plus curl for the download.
set -u

cfg=.vale.ini
detect=0
while getopts c:d opt; do
	case "$opt" in
	c) cfg=$OPTARG ;;
	d) detect=1 ;;
	*) echo "usage: dictionary.sh -d path... | dictionary.sh [-c .vale.ini] en_GB" >&2; exit 2 ;;
	esac
done
shift $((OPTIND - 1))

if [ "$detect" -eq 1 ]; then
	[ $# -gt 0 ] || { echo "dictionary.sh: give at least one path to read" >&2; exit 2; }
	# A fixed list of common pairs. Enough to say which way a corpus leans;
	# not a spell checker.
	us='color|colors|behavior|behaviors|favorite|flavor|honor|labor|center|centers|meter|liter|organize|organized|organization|organizations|recognize|recognized|customize|customized|analyze|analyzed|license|licensed|defense|catalog|dialog|program|programs|canceled|traveled|modeling|labeled|gray|fulfill|enrollment'
	uk='colour|colours|behaviour|behaviours|favourite|flavour|honour|labour|centre|centres|metre|litre|organise|organised|organisation|organisations|recognise|recognised|customise|customised|analyse|analysed|licence|licenced|defence|catalogue|dialogue|programme|programmes|cancelled|travelled|modelling|labelled|grey|fulfil|enrolment'
	files=$(find "$@" -type f \( -name '*.md' -o -name '*.mdx' -o -name '*.markdown' \
		-o -name '*.adoc' -o -name '*.rst' -o -name '*.org' -o -name '*.txt' \) \
		-not -path '*/node_modules/*' -not -path '*/.git/*' -not -path '*/vendor/*' 2>/dev/null)
	[ -n "$files" ] || { echo "dictionary.sh: no prose files under: $*" >&2; exit 2; }
	count() {
		printf '%s\n' "$files" | tr '\n' '\0' | xargs -0 grep -ohwiE "$1" 2>/dev/null | wc -l | tr -d ' '
	}
	top() {
		printf '%s\n' "$files" | tr '\n' '\0' | xargs -0 grep -ohwiE "$1" 2>/dev/null |
			tr '[:upper:]' '[:lower:]' | sort | uniq -c | sort -rn | head -n 4 |
			awk '{ printf "%s%s (%d)", (NR > 1 ? ", " : ""), $2, $1 }'
	}
	a=$(count "$us")
	b=$(count "$uk")
	echo "American spellings: $a   $(top "$us")"
	echo "British spellings:  $b   $(top "$uk")"
	# "program", "license" and "dialog" are also correct British English in
	# computing, so a British corpus still scores some American hits.
	if [ "$a" -eq 0 ] && [ "$b" -eq 0 ]; then
		echo "Leaning: no evidence either way; leave the built-in dictionary."
	elif [ "$b" -gt $((a * 2)) ]; then
		echo "Leaning: British. en_GB would enforce it."
	elif [ "$a" -gt $((b * 2)) ]; then
		echo "Leaning: American. The built-in dictionary is fine; en_US would enforce it."
	else
		echo "Leaning: mixed. Ask which the team intends before enforcing either."
	fi
	exit 0
fi

[ $# -eq 1 ] || { echo "usage: dictionary.sh -d path... | dictionary.sh [-c .vale.ini] <en_GB|fr|de|...>" >&2; exit 2; }
case "$1" in
*[!A-Za-z_-]* | '') echo "dictionary.sh: '$1' is not a language code (en_GB, fr, de, pt-BR)" >&2; exit 2 ;;
esac
# The collection names its folders with a hyphen (en-GB); Hunspell and Vale
# name the files with an underscore (en_GB). American English is plain "en".
name=$(printf '%s' "$1" | tr '-' '_')
src=$(printf '%s' "$1" | tr '_' '-')
case "$name" in
en_US | en) src=en; name=en_US ;;
esac

[ -f "$cfg" ] || { echo "dictionary.sh: no config at $cfg; set Vale up first" >&2; exit 2; }
command -v curl >/dev/null 2>&1 || { echo "dictionary.sh: curl is required" >&2; exit 2; }

sp=$(sed -n 's/^[[:space:]]*StylesPath[[:space:]]*=[[:space:]]*//p' "$cfg" | head -n 1 | tr -d '\r')
[ -n "$sp" ] || { echo "dictionary.sh: $cfg sets no StylesPath" >&2; exit 2; }
case "$sp" in
/* | [A-Za-z]:*) ;;
*) sp=$(dirname "$cfg")/$sp ;;
esac
dest=$sp/config/dictionaries
mkdir -p "$dest"

base=https://raw.githubusercontent.com/wooorm/dictionaries/main/dictionaries/$src
for ext in dic aff; do
	if ! curl -fsSL -o "$dest/$name.$ext.part" "$base/index.$ext"; then
		rm -f "$dest/$name.dic.part" "$dest/$name.aff.part" "$dest/$name.dic"
		echo "dictionary.sh: no dictionary for '$src' at $base" >&2
		echo "See https://github.com/wooorm/dictionaries for the languages it has, or put any" >&2
		echo "Hunspell .dic/.aff pair in $dest yourself." >&2
		exit 1
	fi
	mv "$dest/$name.$ext.part" "$dest/$name.$ext"
done
# A .dic file opens with its entry count, sometimes behind a byte-order mark.
# Anything else is an error page saved under a dictionary's name, and Vale
# would then call every word in the language a misspelling.
entries=$(head -n 1 "$dest/$name.dic" | tr -cd '0-9')
[ -n "$entries" ] || {
	rm -f "$dest/$name.dic" "$dest/$name.aff"
	echo "dictionary.sh: what $base served is not a Hunspell dictionary; nothing installed" >&2
	exit 1
}
curl -fsSL -o "$dest/$name.license.txt" "$base/license" 2>/dev/null || rm -f "$dest/$name.license.txt"

echo "Installed $name.dic and $name.aff in $dest ($entries entries)"
[ -f "$dest/$name.license.txt" ] && echo "Licence: $dest/$name.license.txt"
case "$name" in
en_*) echo "Next: a spelling rule that names $name, and Vale.Spelling = NO. See the vocab skill." ;;
*) echo "Next: Lang = $src on the section for those files. Vale.Spelling then uses this dictionary." ;;
esac
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall