Add multiple configs support and granular cache for icons/images

.claude/EXFIG.toon

@@ -128,6 +128,25 @@ keyDirectories:
   docc: Sources/ExFig/ExFig.docc/
   tests: Tests/
 
+configTypes:
+  IconsConfiguration:
+    purpose: Enum for backward-compatible icons config parsing
+    cases: [single(Icons), multiple([IconsEntry])]
+    platforms: [iOS, Android, Flutter]
+    properties:
+      entries: "[IconsEntry] - unified access to all entries"
+      isMultiple: "Bool - true if array format"
+    decoding: "Try array first, fallback to single object"
+  IconsEntry:
+    purpose: Per-frame icons configuration
+    requiredFields: [format/output (platform-specific)]
+    optionalFields: [figmaFrameName]
+    fallback: "figmaFrameName defaults to common.icons.figmaFrameName or 'Icons'"
+  IconsLoaderConfig:
+    purpose: Sendable struct for IconsLoader frame settings
+    file: Sources/ExFig/Loaders/IconsLoader.swift
+    factoryMethods: [forIOS, forAndroid, forFlutter, defaultConfig]
+
 keyFiles:
   cli: Sources/ExFig/ExFigCommand.swift
   config: Sources/ExFig/Input/Params.swift

.claude/commands/local-review.md

@@ -0,0 +1,95 @@
+---
+description: Run a comprehensive code review of local changes
+allowed-tools: Bash(git *), Grep, Glob, Read, TodoWrite, Task(Explore), mcp__context7__resolve-library-id, mcp__context7__get-library-docs, mcp__sosumi__searchAppleDocumentation, mcp__sosumi__fetchAppleDocumentation
+argument-hint: '[optional: files/directories to review]'
+version: 2.0
+---
+
+# Local Development Review
+
+Comprehensive technical code review analyzing ALL commits in feature branch relative to `origin/develop`.
+
+## Review Scope Decision
+
+```toon
+scope[3]{condition,action}:
+  "specific files/directories provided","review only those paths"
+  "no arguments","review all branch changes vs origin/develop"
+  "module path provided","review module + check cross-module impacts"
+```
+
+## Step 1: Fetch Latest & Identify Changes
+
+```bash
+git fetch origin
+
+# Show all commits in feature branch
+git log origin/develop..HEAD --oneline
+
+# Show all changed files
+git diff origin/develop...HEAD --name-only
+```
+
+If specific files/directories provided via arguments:
+
+```bash
+git diff origin/develop...HEAD --name-only -- [file-paths]
+```
+
+## Step 2: Read Review Guidelines
+
+Read recipe files before flagging issues:
+
+```toon
+recipes[6]{category,path}:
+  "SwiftUI crashes","docs/agents/development/recipes/code-review/swiftui-crash-recipes.md"
+  "Memory management","docs/agents/development/recipes/code-review/memory-management-recipes.md"
+  "Security","docs/agents/development/recipes/code-review/security-recipes.md"
+  "Swift 6 concurrency","docs/agents/development/recipes/code-review/swift6-concurrency-recipes.md"
+  "Architecture violations","docs/agents/development/recipes/code-review/architecture-violation-recipes.md"
+  "Output format","docs/agents/development/recipes/code-review/code-review-output-recipes.md"
+```
+
+## Step 3: Perform Review
+
+Follow @docs/agents/development/code-review.md methodology:
+
+```toon
+methodology[5]{step,action}:
+  1,"Extract context: iOS version, UI framework, architectural layer"
+  2,"Pattern match: verify ALL conditions from detection rules"
+  3,"Assess confidence: >=80% flag, <80% skip"
+  4,"Verify against recipes: read linked recipe files before flagging"
+  5,"Prioritize: CRITICAL -> HIGH -> MEDIUM"
+```
+
+## Step 4: Generate Output
+
+Output format per @docs/agents/development/recipes/code-review/code-review-output-recipes.md:
+
+```toon
+output[3]{section,format,required}:
+  "Code Review Complete","[one line summary]",true
+  "Review Findings","[grouped by priority]",true
+  "No issues","No critical, high, or medium priority issues detected",conditional
+```
+
+Priority format: `emoji + description + file:line + fix action`
+
+```toon
+priorities[3]{level,emoji,examples}:
+  CRITICAL,"red circle","crashes, security, memory leaks"
+  HIGH,"yellow circle","architecture violations, missing error handling"
+  MEDIUM,"green circle","performance, code quality"
+```
+
+## Quality Gate
+
+```toon
+checklist[5]{check,required}:
+  "All changed files reviewed",true
+  "Recipe files consulted for detected issues",true
+  "Confidence >= 80% for all flagged issues",true
+  "Output follows code-review-output-recipes.md format",true
+  "No false positives from partial pattern matches",true
+```

.claude/commands/plan.md

@@ -0,0 +1,144 @@
+---
+description: Break down a task into subtasks with dependencies for parallel execution
+allowed-tools: Read, Glob, Grep, TodoWrite, Task(Explore), Task(Plan)
+argument-hint: <task description>
+version: 1.0
+---
+
+# Task Planning with Dependencies
+
+**Purpose**: Analyze a task and break it down into subtasks with explicit dependencies, enabling parallel execution
+where possible.
+
+## Output Format (TOON-like)
+
+Generate plan in this structured format:
+
+```yaml
+command:
+  name: plan-<task-slug>
+  purpose: <task description>
+  version: 1.0
+
+tasks[N]{id,title,description,depends_on,parallel_group,type}:
+  <id>,<title>,<description>,[deps],[group],[type]
+  ...
+
+execution_order:
+  phase_1:
+    parallel: [task_ids that can run in parallel]
+  phase_2:
+    sequential: [task_id] # build/test tasks
+  phase_3:
+    parallel: [task_ids for fixes if needed]
+  ...
+
+dependency_graph:
+  <task_id>: [list of task_ids this depends on]
+  ...
+```
+
+## Task Types
+
+- `analysis` - Code analysis, research, reading
+- `implementation` - Writing new code
+- `modification` - Changing existing code
+- `build` - Building project/module
+- `test` - Running tests
+- `fix` - Fixing issues found by build/tests
+- `review` - Code review, validation
+
+## Dependency Rules
+
+### Critical Rules for Parallel Execution
+
+1. **Independent tasks** (no shared files/modules) can run in parallel
+2. **Build tasks** MUST wait for ALL parallel implementation tasks to complete
+3. **Test tasks** MUST wait for build to succeed
+4. **Fix tasks** MUST wait for test results
+5. **Tasks modifying same file** MUST be sequential
+
+### Dependency Detection
+
+- Same file modification → sequential
+- Same module modification → sequential (unless different files)
+- Different modules → parallel possible
+- Build depends on → all implementation tasks
+- Test depends on → successful build
+- Fix depends on → test results
+
+## Instructions
+
+1. **Analyze the task**:
+
+   - Read relevant code files mentioned in task
+   - Identify affected modules/files
+   - Detect potential conflicts
+
+2. **Break down into subtasks**:
+
+   - Create atomic, independent subtasks where possible
+   - Identify dependencies between subtasks
+   - Group parallelizable tasks
+
+3. **Generate execution plan**:
+
+   - Phase 1: Parallel analysis/implementation tasks
+   - Phase 2: Build (waits for Phase 1)
+   - Phase 3: Tests (waits for Phase 2)
+   - Phase 4: Fixes if needed (based on Phase 3 results)
+   - Phase 5: Final build/test validation
+
+4. **Output the plan** in TOON format above
+
+## Example Output
+
+```yaml
+command:
+  name: plan-add-analytics-tracking
+  purpose: Add analytics tracking to user profile module
+  version: 1.0
+
+tasks[6]{id,title,description,depends_on,parallel_group,type}:
+  T1,Create analytics service,Implement AnalyticsService protocol,[],G1,implementation
+  T2,Add tracking to ProfileView,Integrate analytics calls,[],G1,implementation
+  T3,Add tracking to SettingsView,Integrate analytics calls,[],G1,implementation
+  T4,Build module,Build UserProfile module,[T1,T2,T3],G2,build
+  T5,Run tests,Execute unit tests,[T4],G3,test
+  T6,Fix issues,Address any test failures,[T5],G4,fix
+
+execution_order:
+  phase_1:
+    parallel: [T1, T2, T3]  # Can run simultaneously - different files
+  phase_2:
+    sequential: [T4]  # Build waits for all implementations
+  phase_3:
+    sequential: [T5]  # Tests wait for build
+  phase_4:
+    conditional: [T6]  # Only if tests fail
+
+dependency_graph:
+  T1: []
+  T2: []
+  T3: []
+  T4: [T1, T2, T3]
+  T5: [T4]
+  T6: [T5]
+
+notes:
+  - T1, T2, T3 modify different files, safe to parallelize
+  - T4 must wait for ALL implementations before building
+  - T6 is conditional - only execute if T5 finds failures
+```
+
+## Validation Checklist
+
+Before finalizing plan, verify:
+
+- [ ] No circular dependencies
+- [ ] Build tasks depend on ALL related implementations
+- [ ] Test tasks depend on successful build
+- [ ] Fix tasks depend on test results
+- [ ] Parallel tasks don't modify same files
+- [ ] All task IDs are unique
+- [ ] dependency_graph matches depends_on fields

.claude/commands/pr-summary.md

@@ -0,0 +1,87 @@
+---
+description: Generate PR summary from branch changes for GitHub
+allowed-tools: Bash(git *), Bash(pbcopy), Read, Glob, Grep, Write
+version: 2.0
+---
+
+# PR Summary Generator
+
+Generate a concise PR summary in English based on branch changes, copy to clipboard.
+
+## Step 1: Gather Context
+
+```bash
+git branch --show-current
+git fetch origin
+git log origin/develop..HEAD --oneline
+git diff origin/develop...HEAD --stat
+```
+
+## Step 2: Extract Task ID
+
+Extract from branch name using pattern `[A-Z]+-[0-9]+`:
+
+```toon
+patterns[4]{branch_example,extracted_id}:
+  "feature/PL-19452-description","PL-19452"
+  "fix/COUR-8147-bug","COUR-8147"
+  "hotfix/PAX-123-critical","PAX-123"
+  "feature/some-description","[manual input required]"
+```
+
+## Step 3: Generate Summary
+
+Write to `/tmp/pr-summary.md`:
+
+```toon
+template{section,format}:
+  Title: "imperative mood, max 50 chars (Add/Fix/Update/Remove)"
+  Task ID: "[{ID}](https://indriver.atlassian.net/browse/{ID})"
+  Description: "1-2 sentences: what changed and why"
+  Changes: "- bullet list of changes"
+```
+
+Template:
+
+```markdown
+## Title
+<imperative mood, max 50 chars>
+
+## Task ID
+[{TASK_ID}](https://indriver.atlassian.net/browse/{TASK_ID})
+
+## Description
+<1-2 sentences>
+
+- <change 1>
+- <change 2>
+```
+
+## Step 4: Copy to Clipboard
+
+```bash
+cat /tmp/pr-summary.md | pbcopy
+```
+
+Inform user: "Copied to clipboard!"
+
+## Output Format
+
+```toon
+sections[4]{name,format,required}:
+  Title,"imperative mood max 50 chars",true
+  "Task ID","markdown link to Jira",true
+  Description,"1-2 sentences explaining what/why",true
+  Changes,"bullet list of specific changes",true
+```
+
+## Quality Gate
+
+```toon
+checklist[5]{check,required}:
+  "Title in imperative mood (not past tense)",true
+  "Task ID extracted and formatted as link",true
+  "English only, formal tone",true
+  "Summary written to /tmp/pr-summary.md",true
+  "Result copied to clipboard via pbcopy",true
+```

.swiftlint.yml

@@ -7,6 +7,7 @@ excluded:
   - Packages
   - Sources/ExFig/Resources/iOSConfig.swift
   - Sources/ExFig/Resources/androidConfig.swift
+  - Sources/ExFig/Resources/flutterConfig.swift
   - Tests/XcodeExportTests/XcodeIconsExporterTests.swift
 
 disabled_rules: