plankit-cli 1.3.1 → 1.4.0

package/templates/dotnet-skills/gemini/dotnet-ef-migration/SKILL.md

@@ -0,0 +1,26 @@
+---
+name: dotnet-ef-migration
+description: Safe Entity Framework Core migration workflow with entity audit, SQL review, and rollback.
+---
+
+# /dotnet-ef-migration
+
+Plan, generate, and review safe Entity Framework Core database migrations.
+
+## Workflow
+1. Audit entity configurations (Fluent API vs DataAnnotations) in target DbContext.
+2. Check for missing indexes, foreign key cascading issues, and property lengths.
+3. Generate migration with a descriptive name (dotnet ef migrations add <Name>).
+4. Generate and inspect raw migration SQL script (dotnet ef migrations script).
+5. Check for destructive schema changes (table drops, column drops, data loss warnings).
+
+## Guardrail
+Never execute irreversible or destructive migrations without explicit user confirmation.
+
+## Options
+- --name <Name>       descriptive migration name
+- --preview-sql       inspect generated SQL before applying
+
+# dotnet-ef-migration
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-endpoint-from-model/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: dotnet-endpoint-from-model
+description: Generate ASP.NET Core Minimal API or Controller endpoints from an entity/model with validation.
+---
+
+# /dotnet-endpoint-from-model
+
+Generate robust CRUD endpoints from a domain model or entity.
+
+## Workflow
+1. Inspect entity/model properties, keys, and navigation properties.
+2. Generate Request/Response DTOs to avoid overposting or leaking domain entities.
+3. Generate endpoint handlers using TypedResults (Ok, Created, NotFound, ValidationProblem).
+4. Configure authorization, route groups, and OpenAPI metadata (.WithName(), .Produces()).
+5. Verify endpoint route constraints.
+
+## Guardrail
+Never expose raw database entities directly to API responses. Always map through DTOs.
+
+## Options
+- --minimal           use ASP.NET Core Minimal APIs (default)
+- --controller        use Controller-based architecture
+- --dto <Type>        use existing DTO type
+
+# dotnet-endpoint-from-model
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-extract-service/SKILL.md

@@ -0,0 +1,26 @@
+---
+name: dotnet-extract-service
+description: Refactor fat services or bloated controllers into isolated CQRS handlers or domain services.
+---
+
+# /dotnet-extract-service
+
+Decompose fat classes into single-responsibility handlers or domain services.
+
+## Workflow
+1. Identify oversized methods or multi-responsibility service classes.
+2. Extract target method into an independent Command/Query handler or domain service.
+3. Define required interfaces and register with Microsoft.Extensions.DependencyInjection.
+4. Update caller references and ensure backward compatibility.
+5. Create unit tests for extracted handler.
+
+## Guardrail
+Preserve existing behavior and contract signatures. Do not introduce breaking API changes.
+
+## Options
+- --class <Name>      source class to refactor
+- --method <Name>     specific method to extract
+
+# dotnet-extract-service
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-feature/SKILL.md

@@ -0,0 +1,27 @@
+---
+name: dotnet-feature
+description: Build an end-to-end vertical slice feature (Command/Query, MediatR, FluentValidation, Endpoint).
+---
+
+# /dotnet-feature
+
+Generate a complete, self-contained feature slice following Vertical Slice Architecture.
+
+## Workflow
+1. Determine feature name, operation type (Command or Query), and target domain.
+2. Generate Request and Response DTO records with immutable properties.
+3. Generate FluentValidation validator with standard validation rules.
+4. Generate MediatR / Wolverine handler with DbContext or repository access.
+5. Expose via Minimal API endpoint or Controller with TypedResults.
+6. Scaffold unit tests and integration tests.
+
+## Guardrail
+Adhere to existing solution conventions (Vertical Slices vs Clean Architecture). Do not introduce unnecessary abstractions.
+
+## Options
+- --type <command|query> operation type
+- --model <Entity>       domain entity to base feature on
+
+# dotnet-feature
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-generate-tests/SKILL.md

@@ -0,0 +1,26 @@
+---
+name: dotnet-generate-tests
+description: Generate xUnit/NUnit tests with FluentAssertions and WebApplicationFactory integration tests.
+---
+
+# /dotnet-generate-tests
+
+Generate comprehensive unit and integration tests for .NET handlers, services, and endpoints.
+
+## Workflow
+1. Inspect target class (Handler, Validator, Service, or Endpoint).
+2. Generate unit tests covering valid input, boundary conditions, and error cases.
+3. Use Moq or NSubstitute for dependencies and FluentAssertions for assertions.
+4. For endpoints, generate WebApplicationFactory integration tests verifying status codes and payloads.
+5. Run tests with dotnet test.
+
+## Guardrail
+Mock only external boundaries (I/O, database, third-party APIs); test real business domain logic.
+
+## Options
+- --framework <xunit|nunit> test framework (default: xunit)
+- --integration             generate WebApplicationFactory integration tests
+
+# dotnet-generate-tests
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-health/SKILL.md

@@ -0,0 +1,25 @@
+---
+name: dotnet-health
+description: Audit .NET solution health: SDK versions, NuGet packages, vulnerabilities, and compiler warnings.
+---
+
+# /dotnet-health
+
+Comprehensive health check for .NET solutions and projects.
+
+## Workflow
+1. Locate .sln and .csproj files, target frameworks (e.g. net8.0, net9.0), and SDK versions.
+2. Check for vulnerable and outdated NuGet packages (dotnet list package --vulnerable).
+3. Audit nullable reference types (<Nullable>enable</Nullable>) and compiler warning configurations.
+4. Output structured markdown diagnostic report with prioritized security and health recommendations.
+
+## Guardrail
+Read-only diagnostic audit. Never modify project files during health check.
+
+## Options
+- --json              output machine-readable JSON metrics
+- --project <path>    restrict audit to a specific project
+
+# dotnet-health
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-migrate-minimal-api/SKILL.md

@@ -0,0 +1,26 @@
+---
+name: dotnet-migrate-minimal-api
+description: Migrate ASP.NET Core controllers to modern Minimal APIs with RouteGroups and typed results.
+---
+
+# /dotnet-migrate-minimal-api
+
+Refactor traditional controllers to modern, high-performance Minimal API route groups.
+
+## Workflow
+1. Inspect controller routes, actions, authorization attributes, and filter pipeline.
+2. Create endpoint extension class (MapXxxEndpoints()) using MapGroup().
+3. Convert ActionResult<T> to TypedResults with explicit return types.
+4. Replace controller DI with endpoint parameter injection ([FromServices], [FromBody]).
+5. Verify route parity and run integration tests.
+
+## Guardrail
+Preserve exact route paths, query parameter names, and HTTP status code behaviors.
+
+## Options
+- --controller <Name> name of the controller to migrate
+- --dry-run           preview changes without modifying files
+
+# dotnet-migrate-minimal-api
+
+Execute the workflow with the target arguments from the request.

package/templates/dotnet-skills/gemini/dotnet-modernize/SKILL.md

@@ -0,0 +1,26 @@
+---
+name: dotnet-modernize
+description: Upgrade legacy C# code to modern C# 12+ / .NET 8+ patterns (primary constructors, pattern matching).
+---
+
+# /dotnet-modernize
+
+Modernize legacy C# syntax using modern C# language features.
+
+## Workflow
+1. Convert block-scoped namespaces to file-scoped namespaces (namespace X;).
+2. Convert classes/records with boilerplate constructors to primary constructors where appropriate.
+3. Adopt collection expressions ([] instead of new List<T>(), Array.Empty<T>()).
+4. Use pattern matching (is not null, switch expressions, property patterns).
+5. Build solution and run test suite to verify zero semantic changes.
+
+## Guardrail
+Only apply safe, semantic-preserving modernizations. Never compromise readability for conciseness.
+
+## Options
+- --project <path>    target specific project
+- --dry-run           preview modernization diffs
+
+# dotnet-modernize
+
+Execute the workflow with the target arguments from the request.

package/templates/vue-commands/codex/component-pattern-audit.md

@@ -0,0 +1,24 @@
+# component-pattern-audit
+> Audit SFC component design, god components, prop drilling, and emits declarations.
+
+# /component-pattern-audit
+
+Evaluate component architecture, complexity, prop drilling, and API design.
+
+## Workflow
+1. Identify god components (> 400 lines, > 15 props, > 10 internal states).
+2. Detect undeclared emits ($emit used in template without defineEmits declaration).
+3. Flag prop drilling deeper than 3 levels where state management is preferable.
+4. Suggest logical component decomposition boundaries.
+
+## Guardrail
+Read-only component evaluation.
+
+## Options
+- --max-lines <n>     line threshold for god component alert (default: 400)
+- --json              output metrics as JSON
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/design-token-discover.md

@@ -0,0 +1,24 @@
+# design-token-discover
+> Discover repeated color literals, typography, spacing, and shadows into design token candidates.
+
+# /design-token-discover
+
+Mine CSS/SCSS and templates for repeated styling values to form design tokens.
+
+## Workflow
+1. Extract all color hex/rgb/hsl values, spacing dimensions, border radii, and font sizes.
+2. Cluster identical and near-identical values.
+3. Suggest semantic token names (--color-primary, --spacing-md, etc.).
+4. Generate candidate token definitions (CSS custom properties, Tailwind theme, or SCSS variables).
+
+## Guardrail
+Do not automatically rewrite styles without approval.
+
+## Options
+- --min-occurrences <n> minimum occurrences to qualify as token (default: 3)
+- --output <file>       output file for candidate tokens
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/design-token-migrate.md

@@ -0,0 +1,24 @@
+# design-token-migrate
+> Migrate hardcoded styles in Vue components to standardized design tokens or utility classes.
+
+# /design-token-migrate
+
+Replace hardcoded colors, spacing, and typography with standardized design tokens.
+
+## Workflow
+1. Match target component styles against the token dictionary.
+2. Replace hardcoded literals with var(--token) or design system utility classes.
+3. Verify visual layout fidelity and verify responsive styling.
+4. Run component tests.
+
+## Guardrail
+Preserve exact visual rendering. Retain fallback values when migrating tokens.
+
+## Options
+- --dry-run           preview style replacements without modifying files
+- --interactive       confirm each token replacement
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/js-to-ts.md

@@ -0,0 +1,25 @@
+# js-to-ts
+> Convert JavaScript Vue SFCs and utility modules to TypeScript with typed props and emits.
+
+# /js-to-ts
+
+Safely migrate JavaScript files and Vue SFCs to TypeScript with strict type definitions.
+
+## Workflow
+1. Rename .js to .ts, or set <script setup lang="ts"> in SFCs.
+2. Convert runtime props to defineProps<{ ... }>() with explicit interfaces.
+3. Convert runtime emits to defineEmits<{ ... }>() with typed payloads.
+4. Add typed interfaces for component state, API responses, and composable returns.
+5. Verify type check with vue-tsc or tsc --noEmit.
+
+## Guardrail
+Do not use 'any' when types can be accurately inferred or declared.
+
+## Options
+- --strict            enforce strict typing without fallback assertions
+- --dry-run           show generated types before applying
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/reactivity-audit.md

@@ -0,0 +1,25 @@
+# reactivity-audit
+> Audit Vue reactivity anti-patterns (destructuring props, missing toRefs, watcher leaks).
+
+# /reactivity-audit
+
+Identify subtle Vue 3 reactivity bugs, lost reactivity, and memory leaks.
+
+## Workflow
+1. Check for destructured defineProps() without toRefs() or toValue() in reactive contexts.
+2. Detect direct prop mutation or nested prop mutations.
+3. Detect watch/watchEffect or window event listeners without cleanup in onUnmounted.
+4. Detect inappropriate mixing of ref and reactive.
+5. Produce fix recommendations with before/after code snippets.
+
+## Guardrail
+Read-only reactivity analysis.
+
+## Options
+- --path <dir>        restrict check to specific directory
+- --fix-suggestions   include detailed refactoring suggestions
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/state-strategy-audit.md

@@ -0,0 +1,23 @@
+# state-strategy-audit
+> Audit Vuex vs Pinia vs composable store usage, identifying leaks and duplicate state.
+
+# /state-strategy-audit
+
+Analyze state management usage across the repository to identify state sprawl and duplication.
+
+## Workflow
+1. Map all stores (Vuex modules, Pinia stores, global reactive composables).
+2. Detect components accessing multiple state paradigms simultaneously.
+3. Flag duplicated domain state (e.g. user info stored in both Vuex and a composable).
+4. Provide migration roadmap for transitioning from Vuex to Pinia.
+
+## Guardrail
+Read-only audit.
+
+## Options
+- --json              output state dependency map as JSON
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-guardrails.md

@@ -0,0 +1,24 @@
+# vue-guardrails
+> Configure ESLint, TypeScript strict rules, and architectural boundaries to prevent regressions.
+
+# /vue-guardrails
+
+Lock in modernization progress by establishing automated linter and compiler guardrails.
+
+## Workflow
+1. Inspect existing ESLint, TypeScript, and Prettier configurations.
+2. Add eslint-plugin-vue rules enforcing <script setup>, defineProps, and defineEmits.
+3. Disallow new Options API usage or deprecated APIs in modernized directories.
+4. Output recommended rule configurations.
+
+## Guardrail
+Never break existing CI builds; introduce strict rules gradually or scoped to modern directories.
+
+## Options
+- --apply             apply recommended configurations to project config files
+- --dry-run           preview configuration changes
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-health.md

@@ -0,0 +1,28 @@
+# vue-health
+> Audit Vue codebase health, tech debt score, Options vs Composition ratio, and modernization risks.
+
+# /vue-health
+
+Comprehensive diagnostic scan of Vue codebase health, architectural consistency, and tech debt.
+
+## Workflow
+1. Discover Vue version, build tooling (Vite/Webpack), language (JS/TS), package manager, and test runner.
+2. Scan SFCs for Options API vs Composition API vs <script setup>.
+3. Scan state management (Vuex vs Pinia vs composables vs local state).
+4. Scan data access (axios in SFCs vs api client vs TanStack Query / useFetch).
+5. Scan styling (global CSS/SCSS vs scoped vs CSS Modules vs Tailwind/utility).
+6. Calculate Health Score (0-100) and classify files into PREFERRED, TRANSITIONAL, and LEGACY.
+7. Output structured markdown diagnostic report with prioritized modernization recommendations.
+
+## Guardrail
+Read-only diagnostic audit. Never modify files during health check.
+
+## Options
+- --json              output machine-readable JSON metrics
+- --min-score <n>     fail if health score is below threshold
+- --path <dir>        restrict analysis to specific directory
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-legacy-map.md

@@ -0,0 +1,25 @@
+# vue-legacy-map
+> Catalog all legacy Vue files, mixins, filters, Options API components, and tech debt hotspots.
+
+# /vue-legacy-map
+
+Build an inventory of legacy code, deprecated APIs, and tech debt hotspots.
+
+## Workflow
+1. Scan for Options API components, this.$ references, mixins, event bus ($on/$emit), and filters.
+2. Detect deprecated CSS deep selectors (::v-deep, /deep/, >>>).
+3. Find legacy Vuex stores and mutation patterns.
+4. Assign migration complexity score (Low, Medium, High, Critical) based on size, coupling, and dependencies.
+5. Generate ranked migration backlog table.
+
+## Guardrail
+Read-only inventory. Do not transform files.
+
+## Options
+- --filter <type>     filter by mixins, vuex, options, or deep-selectors
+- --min-complexity <l> filter by complexity level
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-migration-plan.md

@@ -0,0 +1,24 @@
+# vue-migration-plan
+> Generate a phased, low-risk migration plan for a legacy Vue feature or directory.
+
+# /vue-migration-plan
+
+Formulate an incremental, phased migration plan for a feature slice before code modifications.
+
+## Workflow
+1. Analyze dependencies and call paths of the target directory.
+2. Identify shared state, shared UI primitives, and data access layers.
+3. Divide migration into safe phases: (1) Leaf components, (2) State & data access, (3) Container/Page components, (4) Verification.
+4. Generate PlanKit phase specifications under artifacts/current/<feature>/phases/.
+
+## Guardrail
+Enforce incremental migration over big-bang rewrites. Keep each phase independently testable.
+
+## Options
+- --phases <n>        number of migration phases (default: 3)
+- --feature <name>    name of the feature workspace
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-modernize-slice.md

@@ -0,0 +1,26 @@
+# vue-modernize-slice
+> Perform an end-to-end modernization of a Vue feature slice with validation and guardrails.
+
+# /vue-modernize-slice
+
+Orchestrate complete vertical slice modernization from legacy to target architecture.
+
+## Workflow
+1. Discover feature boundaries, templates, stores, and dependencies.
+2. Generate phased plan with vue-migration-plan.
+3. Modernize components: Options API -> <script setup>, JS -> TS, Vuex -> Pinia.
+4. Apply design tokens with design-token-migrate.
+5. Run vue-tsc, linter, and unit tests to verify zero regressions.
+
+## Guardrail
+Maintain backward compatibility at every step. Do not commit failing builds.
+
+## Options
+- --slice <dir>       path to feature slice directory
+- --dry-run           preview all planned changes
+- --no-tests          skip running tests
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-pattern-map.md

@@ -0,0 +1,24 @@
+# vue-pattern-map
+> Map architectural patterns across the Vue repository for state, data fetching, styling, and components.
+
+# /vue-pattern-map
+
+Map and categorize all coexisting architectural patterns across the Vue codebase.
+
+## Workflow
+1. Identify pattern categories: Component syntax, State management, Data fetching, Styling, Routing, Testing.
+2. Fingerprint each file into its matching pattern archetype.
+3. Cluster patterns and determine: PREFERRED (target), TRANSITIONAL, and LEGACY.
+4. Output a comprehensive pattern directory map and migration complexity ranking.
+
+## Guardrail
+Do not assume repository majority equals target architecture in mixed legacy codebases.
+
+## Options
+- --category <name>   filter by category (state, data, styling, component)
+- --json              output pattern clusters as JSON
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-standardize.md

@@ -0,0 +1,26 @@
+# vue-standardize
+> Refactor a Vue component or composable to conform to the established target architecture.
+
+# /vue-standardize
+
+Refactor a component or composable to align with target conventions with zero breaking changes.
+
+## Workflow
+1. Parse target SFC (script, template, style).
+2. Convert Options API to <script setup> with Composition API.
+3. Replace this.$store with Pinia store or reactive composable.
+4. Preserve public API (props, emits, exposed slots) to avoid breaking parent components.
+5. Verify TypeScript types and run component tests.
+
+## Guardrail
+Never rewrite business logic. Always preserve component prop/event contracts and visual styling.
+
+## Options
+- --dry-run           preview changes without writing
+- --no-tests          skip running verification tests
+- --interactive       confirm transformations step by step
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-style-audit.md

@@ -0,0 +1,24 @@
+# vue-style-audit
+> Audit styling patterns, global vs scoped CSS/SCSS, deep selectors, and hardcoded values.
+
+# /vue-style-audit
+
+Analyze CSS/SCSS architecture, scoping violations, and hardcoded values across Vue components.
+
+## Workflow
+1. Catalog all <style> tags: scoped vs unscoped, lang (css, scss, less), and CSS Modules.
+2. Identify leaky global styles and deprecated deep selectors (::v-deep, /deep/).
+3. Detect hardcoded hex colors, magic numbers in margins/padding, font sizes, and z-indexes.
+4. Generate style report highlighting candidates for design token extraction.
+
+## Guardrail
+Read-only style audit.
+
+## Options
+- --json              output style metrics as JSON
+- --path <dir>        restrict audit to specific directory
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/codex/vue-target-architecture.md

@@ -0,0 +1,24 @@
+# vue-target-architecture
+> Define or inspect the target architectural conventions for modernizing the Vue application.
+
+# /vue-target-architecture
+
+Establish and document the canonical target architecture for the repository.
+
+## Workflow
+1. Inspect existing modern slices or prompt for target conventions.
+2. Define target standards: Vue 3, <script setup>, TypeScript strictness, Pinia stores, styling foundation, testing framework.
+3. Classify patterns into PREFERRED, TRANSITIONAL, and LEGACY.
+4. Record target architectural profile in artifacts/VUE_TARGET.md for PlanKit commands to reference.
+
+## Guardrail
+Require explicit target decisions before broad refactoring. Never guess without evidence.
+
+## Options
+- --interactive       guide through target architecture decisions
+- --save              persist target architecture to artifacts
+
+## Execution
+1. Run Vue context discovery against the repository.
+2. Apply the workflow above with the provided target.
+3. Report files created/modified and patterns standardized.

package/templates/vue-commands/cursor/component-pattern-audit.md

@@ -0,0 +1,25 @@
+---
+description: Audit SFC component design, god components, prop drilling, and emits declarations.
+globs: **/*.vue, **/*.ts, **/*.js, **/*.scss, **/*.css
+---
+
+# component-pattern-audit
+
+# /component-pattern-audit
+
+Evaluate component architecture, complexity, prop drilling, and API design.
+
+## Workflow
+1. Identify god components (> 400 lines, > 15 props, > 10 internal states).
+2. Detect undeclared emits ($emit used in template without defineEmits declaration).
+3. Flag prop drilling deeper than 3 levels where state management is preferable.
+4. Suggest logical component decomposition boundaries.
+
+## Guardrail
+Read-only component evaluation.
+
+## Options
+- --max-lines <n>     line threshold for god component alert (default: 400)
+- --json              output metrics as JSON
+
+Follow this workflow whenever asked to modernize, standardize, or audit Vue components in this repository.

package/templates/vue-commands/cursor/design-token-discover.md

@@ -0,0 +1,25 @@
+---
+description: Discover repeated color literals, typography, spacing, and shadows into design token candidates.
+globs: **/*.vue, **/*.ts, **/*.js, **/*.scss, **/*.css
+---
+
+# design-token-discover
+
+# /design-token-discover
+
+Mine CSS/SCSS and templates for repeated styling values to form design tokens.
+
+## Workflow
+1. Extract all color hex/rgb/hsl values, spacing dimensions, border radii, and font sizes.
+2. Cluster identical and near-identical values.
+3. Suggest semantic token names (--color-primary, --spacing-md, etc.).
+4. Generate candidate token definitions (CSS custom properties, Tailwind theme, or SCSS variables).
+
+## Guardrail
+Do not automatically rewrite styles without approval.
+
+## Options
+- --min-occurrences <n> minimum occurrences to qualify as token (default: 3)
+- --output <file>       output file for candidate tokens
+
+Follow this workflow whenever asked to modernize, standardize, or audit Vue components in this repository.