@harness-engineering/cli 1.7.0 → 1.8.0

package/dist/agents/skills/gemini-cli/harness-soundness-review/skill.yaml

@@ -0,0 +1,48 @@
+name: harness-soundness-review
+version: "1.0.0"
+description: Deep soundness analysis of specs and plans with auto-fix and convergence loop
+cognitive_mode: meticulous-verifier
+triggers:
+  - manual
+platforms:
+  - claude-code
+  - gemini-cli
+tools:
+  - Bash
+  - Read
+  - Write
+  - Edit
+  - Glob
+  - Grep
+cli:
+  command: harness skill run harness-soundness-review
+  args:
+    - name: path
+      description: Project root path
+      required: false
+    - name: mode
+      description: Review mode — "spec" for spec soundness or "plan" for plan soundness
+      required: true
+mcp:
+  tool: run_skill
+  input:
+    skill: harness-soundness-review
+    path: string
+type: rigid
+phases:
+  - name: check
+    description: Run all checks for the current mode and classify findings
+    required: true
+  - name: fix
+    description: Auto-fix inferrable issues and log changes
+    required: true
+  - name: converge
+    description: Re-run checks, compare issue counts, loop or stop
+    required: true
+  - name: surface
+    description: Present remaining issues to user for resolution
+    required: true
+state:
+  persistent: false
+  files: []
+depends_on: []

package/dist/agents/skills/gemini-cli/harness-test-advisor/SKILL.md

@@ -13,8 +13,23 @@
 
 ## Prerequisites
 
-A knowledge graph must exist at `.harness/graph/`. Run `harness scan` if no graph is available.
-If the graph exists but code has changed since the last scan, re-run `harness scan` first — stale graph data leads to inaccurate results.
+A knowledge graph at `.harness/graph/` enables full analysis. If no graph exists,
+the skill uses static analysis fallbacks (see Graph Availability section).
+Run `harness scan` to enable graph-enhanced analysis.
+
+### Graph Availability
+
+Before starting, check if `.harness/graph/graph.json` exists.
+
+**If graph exists:** Check staleness — compare `.harness/graph/metadata.json`
+scanTimestamp against `git log -1 --format=%ct` (latest commit timestamp).
+If graph is more than 10 commits behind (`git log --oneline <scanTimestamp>..HEAD | wc -l`),
+run `harness scan` to refresh before proceeding. (Staleness sensitivity: **Medium**)
+
+**If graph exists and is fresh (or refreshed):** Use graph tools as primary strategy.
+
+**If no graph exists:** Output "Running without graph (run `harness scan` to
+enable full analysis)" and use fallback strategies for all subsequent steps.
 
 ## Process
 
@@ -43,6 +58,20 @@ For each changed file, use graph traversal to find test files:
 
 3. **Co-change tests**: Check `co_changes_with` edges for test files that historically change alongside the modified files.
 
+#### Fallback (without graph)
+
+When no graph is available, use naming conventions, import parsing, and git history:
+
+1. **Tier 1 — Filename convention matching**: For each changed file `foo.ts`, search for:
+   - `foo.test.ts`, `foo.spec.ts` (same directory)
+   - `__tests__/foo.ts`, `__tests__/foo.test.ts`
+   - Test files in a parallel `tests/` directory mirroring the source path
+2. **Tier 2 — Import-linked tests**: Parse test files' import statements (grep for `import.*from` in `*.test.*` and `*.spec.*` files). If a test file imports the changed file, it belongs in Tier 2 (if not already in Tier 1).
+3. **Tier 3 — Co-change correlated tests**: Use `git log --format="%H" --name-only` to find test files that frequently change in the same commit as the target file. Files that co-change in >2 commits are co-change correlated.
+4. **Rank**: Tier 1 = direct filename match, Tier 2 = import-linked tests, Tier 3 = co-change correlated tests. Output the same tiered format as the graph version.
+
+> Fallback completeness: ~80% — naming conventions and imports catch most mappings; misses dynamic imports and indirect coverage.
+
 ### Phase 3: PRIORITIZE — Rank and Generate Commands
 
 Organize tests into three tiers:
@@ -85,7 +114,7 @@ npx vitest run tests/services/auth.test.ts tests/types/user.test.ts tests/routes
 
 ## Harness Integration
 
-- **`harness scan`** — Must run before this skill to ensure graph is current.
+- **`harness scan`** — Recommended before this skill for full graph-enhanced analysis. If graph is missing, skill uses naming convention and import parsing fallbacks.
 - **`harness validate`** — Run after acting on findings to verify project health.
 - **Graph tools** — This skill uses `query_graph`, `get_impact`, and `get_relationships` MCP tools.
 
@@ -95,7 +124,7 @@ npx vitest run tests/services/auth.test.ts tests/types/user.test.ts tests/routes
 - Executable run commands generated for quick and full test runs
 - Coverage gaps flagged for changed files with no test coverage
 - Report follows the structured output format
-- All findings are backed by graph query evidence, not heuristics
+- All findings are backed by graph query evidence (with graph) or systematic static analysis (without graph)
 
 ## Examples
 
@@ -122,8 +151,8 @@ Output:
 
 ## Gates
 
-- **No advice without graph.** If no graph exists, fall back to: "Run all tests in the same directory as changed files."
-- **Always include Tier 1.** Direct test coverage is non-negotiable — always recommend running these.
+- **Graph preferred, fallback available.** If no graph exists, use naming conventions, import parsing, and git co-change analysis to identify relevant tests. Do not stop — produce the best test selection possible.
+- **Always include Tier 1.** Direct test coverage is non-negotiable — always recommend running these (whether found via graph or naming conventions).
 
 ## Escalation
 

package/dist/agents/skills/shared/i18n-knowledge/accessibility/intersection.yaml

@@ -0,0 +1,142 @@
+description: "Rules at the intersection of i18n and accessibility — ensuring localized content remains accessible across languages, scripts, and text directions"
+
+rules:
+  - name: "lang attribute on html element"
+    category: "lang-tags"
+    wcag_criteria: ["3.1.1"]
+    description: "The <html> element must have a valid lang attribute matching the page's primary language"
+    detect:
+      method: "Check for lang attribute on <html> element; verify it matches a valid BCP 47 tag and corresponds to the page content language"
+    reason: "Screen readers use the lang attribute to select the correct pronunciation engine. Without it, an English screen reader will mispronounce French content."
+    fix: "Set <html lang='fr'> for French content, <html lang='ar' dir='rtl'> for Arabic, etc. Update dynamically when locale changes."
+    severity: error
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "lang attribute on inline language switches"
+    category: "lang-tags"
+    wcag_criteria: ["3.1.2"]
+    description: "Inline content in a different language must be wrapped with a lang attribute"
+    detect:
+      method: "Check for foreign-language words or phrases without a lang attribute on their container"
+    reason: "Screen readers will pronounce foreign words using the page's primary language rules, making them unintelligible."
+    fix: "Wrap language switches: <span lang='fr'>bonjour</span> within English content, <span lang='en'>API</span> within Japanese content."
+    severity: warning
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "dir attribute for RTL content"
+    category: "bidi-a11y"
+    wcag_criteria: ["1.3.2"]
+    description: "RTL content must have dir='rtl' set on the appropriate container element"
+    detect:
+      method: "Check for Arabic, Hebrew, or other RTL script content without a dir attribute on its container"
+    reason: "Without dir='rtl', screen readers may read RTL text in the wrong order, and visual rendering will be incorrect."
+    fix: "Set dir='rtl' on containers with RTL content. For the full page: <html lang='ar' dir='rtl'>. For inline: <span dir='rtl'>."
+    severity: error
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "dir=auto on user-generated content"
+    category: "bidi-a11y"
+    wcag_criteria: ["1.3.2"]
+    description: "User-generated content containers should use dir='auto' to detect text direction automatically"
+    detect:
+      method: "Check for user-generated content areas (comments, messages, reviews) without dir='auto'"
+    reason: "User content may be in any language. Without dir='auto', an Arabic comment in an English page will render incorrectly."
+    fix: "Set dir='auto' on elements that display user-generated content: <p dir='auto'>{userComment}</p>"
+    severity: warning
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Font size scaling for complex scripts"
+    category: "script-sizing"
+    wcag_criteria: ["1.4.4"]
+    description: "Complex scripts (Devanagari, Thai, Arabic, CJK) require larger minimum font sizes for legibility"
+    detect:
+      method: "Check minimum font sizes when content language uses complex scripts; flag sizes below recommended minimums"
+    reason: "Complex scripts have more visual detail per character. Arabic letters reshape by position. Thai stacks diacritics vertically. Below 12-14px, these become illegible."
+    fix: "Set minimum font sizes: 12px for CJK, 13px for Arabic, 14px for Devanagari and Thai. Increase line-height to 1.6-1.8 for scripts with stacking marks."
+    severity: warning
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "Screen reader pronunciation of numbers and dates"
+    category: "screen-readers"
+    wcag_criteria: ["1.3.1"]
+    description: "Numbers and dates must use semantic markup so screen readers pronounce them correctly per locale"
+    detect:
+      method: "Check for visually formatted numbers/dates that lack semantic markup (time element, appropriate ARIA)"
+    reason: "A screen reader seeing '03/04/2025' cannot determine if it is March 4 or April 3 without semantic context. '1.234' could be one-point-two-three-four or one-thousand-two-hundred-thirty-four."
+    fix: "Use <time datetime='2025-03-04'>March 4, 2025</time> for dates. Use aria-label for ambiguous number formats."
+    severity: warning
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "ARIA labels must be translated"
+    category: "screen-readers"
+    wcag_criteria: ["4.1.2"]
+    description: "All ARIA labels, descriptions, and live region content must go through the translation pipeline"
+    detect:
+      method: "Check for hardcoded English strings in aria-label, aria-description, aria-roledescription attributes"
+    reason: "Untranslated ARIA labels make the interface inaccessible to screen reader users in non-English locales."
+    fix: "Wrap ARIA attributes in translation functions: aria-label={t('nav.menu')} instead of aria-label='Menu'"
+    severity: error
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "hreflang attribute on alternate-language links"
+    category: "lang-tags"
+    wcag_criteria: ["3.1.1"]
+    description: "Links to alternate-language versions of a page must have hreflang attributes"
+    detect:
+      method: "Check for <link rel='alternate'> elements without hreflang, or language switcher links without hreflang"
+    reason: "hreflang helps search engines serve the correct language version and assists screen readers in announcing the target language."
+    fix: "Add hreflang to all alternate links: <link rel='alternate' hreflang='es' href='/es/page'>. Include x-default for the default version."
+    severity: warning
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Translated alt text for images"
+    category: "screen-readers"
+    wcag_criteria: ["1.1.1"]
+    description: "Image alt text must be translated along with all other user-facing content"
+    detect:
+      method: "Check for img alt attributes containing English text when the page language is non-English"
+    reason: "Untranslated alt text is meaningless to screen reader users who do not speak the source language."
+    fix: "Include all alt attributes in the translation pipeline: <img alt={t('hero.alt')} /> instead of hardcoded text"
+    severity: error
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Keyboard navigation in RTL layouts"
+    category: "bidi-a11y"
+    wcag_criteria: ["2.4.3"]
+    description: "Tab order must follow visual order in RTL layouts — right-to-left, top-to-bottom"
+    detect:
+      method: "Check tab order in RTL layouts; verify it matches the visual reading order (right to left)"
+    reason: "If tab order remains LTR in an RTL layout, keyboard users navigate against the visual flow, causing confusion."
+    fix: "Use CSS logical properties for layout; avoid manual tabindex that assumes LTR order; test tab navigation with RTL locale"
+    severity: error
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error

package/dist/agents/skills/shared/i18n-knowledge/anti-patterns/encoding.yaml

@@ -0,0 +1,67 @@
+description: "Encoding anti-patterns — common mistakes in character encoding, Unicode handling, and text measurement that break with non-ASCII content"
+
+patterns:
+  - name: "Assuming 1 character equals 1 byte"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for buffer allocation or data storage sized by string.length for strings that may contain non-ASCII"
+      context: "Buffer.alloc(str.length) or database VARCHAR columns sized by character count for UTF-8 data"
+    reason: "UTF-8 encodes characters in 1-4 bytes. 'hello' is 5 bytes but 'こんにちは' is 15 bytes. Buffer overflow or truncation results."
+    instead: "Use Buffer.byteLength(str, 'utf8') for byte calculations; size database columns for byte length not character count"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "String length not equal to visual width"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for string.length used to estimate visual display width"
+      context: "Layout calculations or column alignment using .length instead of visual width measurement"
+    reason: "CJK characters are double-width visually but .length counts them as 1. Emoji may be multiple code points but one visual glyph."
+    instead: "Use Intl.Segmenter for grapheme counting; use canvas.measureText() or terminal wcwidth for visual width"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Missing UTF-8 BOM handling"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for file reading that does not strip BOM (U+FEFF) from the beginning of text files"
+      context: "Reading CSV, JSON, or translation files that may have been edited in Windows tools that add BOM"
+    reason: "BOM at the start of a file can break JSON parsing, CSV parsing, and string comparison. Windows Notepad adds BOM by default."
+    instead: "Strip BOM when reading files: content.replace(/^\\uFEFF/, '') or use a BOM-aware file reader"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Emoji and ZWJ sequence handling"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for string operations that split or truncate without grapheme cluster awareness"
+      context: "Array.from(str) or str.split('') used to process strings containing emoji"
+    reason: "Family emoji (👨‍👩‍👦) is 5 code points joined by ZWJ. Splitting by code point produces broken fragments."
+    instead: "Use Intl.Segmenter(locale, { granularity: 'grapheme' }) to iterate over visual characters safely"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Filename and URL encoding not handling non-ASCII"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for file path or URL construction that does not encode non-ASCII characters"
+      context: "Path concatenation with user input or locale-specific filenames without encodeURIComponent"
+    reason: "Non-ASCII filenames and URL segments must be percent-encoded. Unencoded characters cause 404 errors and security issues."
+    instead: "Use encodeURIComponent() for URL segments; use encodeURI() for full URLs; normalize filenames to ASCII-safe slugs"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error

package/dist/agents/skills/shared/i18n-knowledge/anti-patterns/formatting.yaml

@@ -0,0 +1,106 @@
+description: "Formatting anti-patterns — common mistakes in number, date, currency, phone, and address formatting across locales"
+
+patterns:
+  - name: "Date formatting without locale"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for toLocaleDateString() without locale argument or manual date string construction"
+      context: "new Date().toLocaleDateString() without first arg, or month + '/' + day + '/' + year concatenation"
+    reason: "Without explicit locale, the runtime's default locale is used — which may differ from the user's locale on servers or CI."
+    instead: "Always pass locale: new Intl.DateTimeFormat(userLocale, options).format(date)"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "Number formatting with hardcoded separators"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for regex or replace operations that insert commas or periods as number separators"
+      context: "value.toFixed(2).replace('.', ',') or regex-based thousand separator insertion"
+    reason: "Decimal separator is comma in most of Europe (1.234,56), period in US/UK (1,234.56), and momayyez in Arabic."
+    instead: "Use Intl.NumberFormat(locale).format(value) for all number display"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Currency display with hardcoded symbol"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for string concatenation with currency symbols: '$' + amount or amount + ' EUR'"
+      context: "Currency symbol prepended or appended to a formatted number"
+    reason: "Symbol position varies: $100 (en-US), 100 $ (fr-FR), 100$ (pt-BR). Some currencies use code (CHF) not symbol."
+    instead: "Use Intl.NumberFormat(locale, { style: 'currency', currency: code }).format(amount)"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Phone number formatting without libphonenumber"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for regex-based phone number formatting or validation"
+      context: "Phone regex like /^\\d{10}$/ or manual formatting with dashes and parentheses"
+    reason: "Phone number formats vary enormously by country. UK: 020 7946 0958. US: (202) 555-0123. Japan: 03-1234-5678."
+    instead: "Use google-libphonenumber for parsing, validation, and formatting; store in E.164 format"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Address formatting with hardcoded field order"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for address string templates that assume a fixed field order"
+      context: "Template like `${street}, ${city}, ${state} ${zip}` used for all countries"
+    reason: "Japan: postal code first, then prefecture, city, street. Germany: street, then postal code + city (same line). Formats vary worldwide."
+    instead: "Use Google's libaddressinput or i18n-postal-address for country-specific address formatting"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Percentage formatting assumptions"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for percentage display constructed by appending '%' to a number"
+      context: "value + '%' or `${value}%` without locale-aware formatting"
+    reason: "Some locales put a space before % (French: 25 %). Turkish uses %25 (prefix). Arabic may use the Arabic percent sign."
+    instead: "Use Intl.NumberFormat(locale, { style: 'percent' }).format(value / 100)"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Hardcoded decimal precision"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for .toFixed(2) applied to currency amounts"
+      context: "Universal 2-decimal-place formatting for all currencies"
+    reason: "JPY and KRW have 0 decimal places. BHD and KWD have 3. Using 2 for all misrepresents amounts."
+    instead: "Use Intl.NumberFormat with currency option which auto-selects correct precision per currency code"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Time zone display without IANA identifier"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for timezone abbreviations (EST, PST, CET) in user-facing display"
+      context: "Displaying 'EST' or 'PST' without the full IANA timezone identifier"
+    reason: "Timezone abbreviations are ambiguous (CST = Central Standard Time or China Standard Time) and not universally understood."
+    instead: "Use IANA identifiers (America/New_York) for storage; display user-friendly names via Intl.DateTimeFormat with timeZoneName option"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn

package/dist/agents/skills/shared/i18n-knowledge/anti-patterns/layout.yaml

@@ -0,0 +1,80 @@
+description: "Layout anti-patterns — common mistakes in spatial design, text containers, and directional assumptions that break in localized contexts"
+
+patterns:
+  - name: "Fixed-width containers for text"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for fixed pixel widths on containers that hold translated text"
+      context: "Buttons, labels, or cards with hardcoded width in px that hold user-facing text"
+    reason: "German expands 30-35% over English. Finnish expands up to 40%. Fixed-width containers cause text overflow or truncation."
+    instead: "Use min-width/max-width with flexible layouts; test with the longest target language (typically German or Finnish)"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "Hardcoded CSS left/right instead of logical properties"
+    severity: warning
+    scope: web
+    detect:
+      method: "Check for CSS properties using physical directions: margin-left, padding-right, text-align: left, float: left"
+      context: "Layout CSS that uses physical direction properties instead of logical equivalents"
+    reason: "Physical left/right breaks in RTL layouts. margin-left becomes the wrong side for Arabic and Hebrew users."
+    instead: "Use CSS logical properties: margin-inline-start, padding-inline-end, text-align: start, float: inline-start"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Icon placement assumptions"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for icons hardcoded to left or right side of text"
+      context: "Arrow icons, action icons, or navigation indicators with fixed directional placement"
+    reason: "Left-side icons in LTR become wrong-side icons in RTL. But some icons should NOT mirror (checkmarks, media play buttons)."
+    instead: "Use logical placement (inline-start/end); create an icon exception list for icons that should not mirror in RTL"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Text truncation without dir attribute"
+    severity: warning
+    scope: web
+    detect:
+      method: "Check for text-overflow: ellipsis without corresponding dir attribute on the container"
+      context: "Truncated text elements that may contain RTL content"
+    reason: "Ellipsis appears at the wrong end for RTL text without dir attribute. Bidi truncation shows '...مرحبا' instead of 'مرحبا...'"
+    instead: "Set dir attribute on text containers; use CSS direction property in conjunction with text-overflow"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Fixed height on text containers"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for fixed pixel heights on containers that hold multi-line translated text"
+      context: "Cards, tooltips, or modals with hardcoded height that hold translatable content"
+    reason: "Translated text may wrap to more lines due to expansion. Thai and Devanagari need more vertical space for diacritics."
+    instead: "Use min-height instead of height; increase line-height for scripts with stacking diacritics (Thai: 1.8, Devanagari: 1.6)"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Images or graphics containing embedded text"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for images, SVGs, or canvas elements with text baked into the visual"
+      context: "Hero images, infographics, or icons with embedded text labels"
+    reason: "Text in images cannot be translated, searched, or read by screen readers. Each locale requires a new image asset."
+    instead: "Overlay text on images using CSS positioning; use SVG with <text> elements that can reference translation keys"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error

package/dist/agents/skills/shared/i18n-knowledge/anti-patterns/pluralization.yaml

@@ -0,0 +1,80 @@
+description: "Pluralization anti-patterns — common mistakes in handling plural forms across languages with varying CLDR plural categories"
+
+patterns:
+  - name: "Binary plural logic"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for ternary or if/else that only handles singular and plural (two forms)"
+      context: "count === 1 ? singular : plural or count > 1 ? plural : singular"
+    reason: "Many languages need more than two forms. Arabic has 6 (zero, one, two, few, many, other). Polish has 4. Russian has 4. Binary logic covers only English-like languages."
+    instead: "Use ICU MessageFormat plural syntax with all needed CLDR categories: {count, plural, zero {...} one {...} two {...} few {...} many {...} other {...}}"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Hardcoded English plural rules applied to all locales"
+    severity: error
+    scope: all
+    detect:
+      method: "Check for plural functions that add 's' suffix or check count === 1"
+      context: "Functions like pluralize(word, count) that append 's' or select between two English forms"
+    reason: "English plural rule (add 's') is specific to English. Most languages have completely different plural morphology."
+    instead: "Delegate pluralization entirely to the i18n framework which implements CLDR plural rules per locale"
+    strictness:
+      permissive: warn
+      standard: error
+      strict: error
+
+  - name: "Missing CLDR plural categories for target locales"
+    severity: error
+    scope: all
+    detect:
+      method: "Check translation files for missing plural category keys compared to what CLDR requires for each locale"
+      context: "Arabic translation file with only 'one' and 'other' forms instead of all 6 required forms"
+    reason: "Missing plural forms cause fallback to 'other' which may be grammatically incorrect for specific counts."
+    instead: "Audit translation files against CLDR plural rules for each target locale; ensure all required categories are present"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: error
+
+  - name: "Ordinal pluralization not handled"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for ordinal display (1st, 2nd, 3rd) constructed manually or with English-only rules"
+      context: "Appending 'st', 'nd', 'rd', 'th' suffixes based on English ordinal rules"
+    reason: "Ordinal rules vary by locale. English: 1st, 2nd, 3rd. French: 1er, 2e. German: 1., 2. Some languages have no written ordinal form."
+    instead: "Use Intl.PluralRules(locale, { type: 'ordinal' }) to select the correct ordinal form per locale"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Plural forms embedded in UI code instead of message format"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for conditional rendering of plural text in component code rather than translation files"
+      context: "JSX like {count === 1 ? <span>1 item</span> : <span>{count} items</span>}"
+    reason: "Plural logic in code bypasses the translation pipeline. Translators cannot adapt plural forms for their locale."
+    instead: "Move plural selection to translation files using ICU or framework plural syntax; render with a single translation key"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn
+
+  - name: "Count-based display without using plural-aware formatter"
+    severity: warning
+    scope: all
+    detect:
+      method: "Check for number + noun patterns constructed without plural formatter"
+      context: "Displaying '5 file(s)' or using parenthetical plural as a workaround"
+    reason: "The '(s)' workaround is not grammatical in any language. It looks unprofessional and confuses screen readers."
+    instead: "Use proper plural formatting: t('files', { count: 5 }) with locale-specific plural forms in translation files"
+    strictness:
+      permissive: info
+      standard: warn
+      strict: warn