CLI — Workflows & reporting
sync
Synopsis: ai-i18n-tools sync [options]
Extract (if enabled), then UI translation, then translate-svg when features.translateSVG and config.svg are set, then documentation translation, then translate-json when features.translateJson and json[] are set — unless skipped with --no-ui, --no-svg, --no-docs, or --no-json.
Key options: -l, -p / -f, --dry-run, -j, -b, --force, --force-update, --check-cache, --no-ui, --no-svg, --no-docs, --no-json
--force is forwarded to the UI and SVG steps as well as docs/JSON; --force-update applies to docs, JSON, and SVG (not UI). --check-cache is forwarded to docs, JSON, and SVG: it re-validates cached segments for locales with an enforced native script even when file tracking would skip. Docs phase also forwards --emphasis-placeholders (same meaning as translate-docs). Global --debug-failed writes FAILED-TRANSLATION logs under cacheDir for each discarded model attempt (including SVG/docs script fallbacks), not only when every model in the chain fails. --prompt-format is not a sync flag; docs and JSON steps use the built-in default (json-array).
status
Synopsis: ai-i18n-tools status [--max-columns <n>]
When features.translateUIStrings is on, prints UI coverage per locale (Translated / Missing / Total). Then prints markdown translation status per file × locale (no --locale filter; locales come from config). When features.translateJson is on and json[] is configured, also prints JSON bundle status per block. Large locale lists are split into repeated tables of up to n locale columns (default 9) so lines stay narrow in the terminal.
Key options: --max-columns
statistics
Synopsis: ai-i18n-tools statistics [--max-columns <n>]
Print documentation cache and strings.json statistics (same aggregates as Translation Dashboard → Statistics). --max-columns: max locale columns per model × locale table (default 6).
Key options: --max-columns
See also: Dashboard statistics
usage
Synopsis: ai-i18n-tools usage [--since <when>] [--provider <name>] [--model <id>] [--operation <name>] [-l <code>] [--outcome accepted|discarded] [--clear] [--older-than <when>] [--dry-run]
Print recorded model API-call statistics (calls, tokens, and a single USD cost). Cost is the provider's usage.cost when present, otherwise the amount from providers.<name>.modelPricing or the provider-wide providers.<name>.pricing default (stored on new calls; applied at report time for older rows that have no stored cost). Same aggregates as Translation Dashboard → Usage & costs. Detail rows older than seven UTC calendar days are rolled into monthly api_totals; reports combine both tables. --since accepts YYYY-MM-DD, a duration (30m, 1h, 6h, 12h, 24h, 7d, 30d), or a calendar-month window (1mo, 2mo, 3mo). --clear deletes detail rows and monthly totals (--older-than is 1mo, 2mo, 3mo, 6mo, 1y, or all; --dry-run reports the count without deleting).
Key options: --since, --provider, --model, --operation, -l / --locale, --outcome, --clear, --older-than, --dry-run
See also: Dashboard usage & costs