dandi · yarikoptic · May 5, 2026 · May 5, 2026
diff --git a/.lad/.copilot-instructions.md b/.lad/.copilot-instructions.md
diff --git a/.lad/CLAUDE.md b/.lad/CLAUDE.md
diff --git a/.lad/LAD_RECIPE.md b/.lad/LAD_RECIPE.md
diff --git a/.lad/LICENSE.md b/.lad/LICENSE.md
diff --git a/.lad/README.md b/.lad/README.md
@@ -6,21 +6,21 @@ LAD enables **systematic feature development** and **enterprise-grade test quali
 
 ## ✨ What's New in 2025
 
-🔬 **Enhanced Test Quality Framework** — Achieve 90%+ test success through systematic PDCA cycles  
-🎯 **Industry Standards Compliance** — Research software + Enterprise + IEEE validation  
-📊 **Session Continuity** — Seamless interruption/resumption across multiple sessions  
-⚡ **Real-World Insights** — Based on 50+ LAD implementations in research software  
+🔬 **Enhanced Test Quality Framework** — Achieve 90%+ test success through systematic PDCA cycles
+🎯 **Industry Standards Compliance** — Research software + Enterprise + IEEE validation
+📊 **Session Continuity** — Seamless interruption/resumption across multiple sessions
+⚡ **Real-World Insights** — Based on 50+ LAD implementations in research software
 
 ## Features
 
-✅ **Test-driven development** with atomic task breakdowns  
-✅ **Systematic test improvement** with PDCA methodology  
-✅ **Component-aware testing** (integration for APIs, unit for business logic)  
-✅ **Multi-level documentation** with collapsible sections  
-✅ **NumPy-style docstrings** enforced throughout  
-✅ **Session continuity** with TodoWrite progress tracking  
-✅ **GitHub Flow** with automated PR creation/cleanup  
-✅ **Agent autonomy** with diff approval workflow  
+✅ **Test-driven development** with atomic task breakdowns
+✅ **Systematic test improvement** with PDCA methodology
+✅ **Component-aware testing** (integration for APIs, unit for business logic)
+✅ **Multi-level documentation** with collapsible sections
+✅ **NumPy-style docstrings** enforced throughout
+✅ **Session continuity** with TodoWrite progress tracking
+✅ **GitHub Flow** with automated PR creation/cleanup
+✅ **Agent autonomy** with diff approval workflow
 
 ## Choose Your Workflow
 
@@ -118,10 +118,10 @@ Claude: I'll use the enhanced test quality framework to systematically improve y
 
 ## Documentation
 
-📖 **[LAD_RECIPE.md](LAD_RECIPE.md)** — Complete step-by-step workflow guide  
-🚀 **[Claude Code prompts](claude_prompts/)** — 7-phase autonomous workflow  
-🛠️ **[Copilot Agent prompts](copilot_prompts/)** — Function-based autonomous workflow  
-🔬 **Enhanced Test Quality** — 4-phase systematic improvement framework  
+📖 **[LAD_RECIPE.md](LAD_RECIPE.md)** — Complete step-by-step workflow guide
+🚀 **[Claude Code prompts](claude_prompts/)** — 7-phase autonomous workflow
+🛠️ **[Copilot Agent prompts](copilot_prompts/)** — Function-based autonomous workflow
+🔬 **Enhanced Test Quality** — 4-phase systematic improvement framework
 
 ## Requirements
 
@@ -130,7 +130,7 @@ Claude: I'll use the enhanced test quality framework to systematically improve y
 - Python 3.11+
 - Git repository
 
-### For Copilot Agent Workflow  
+### For Copilot Agent Workflow
 - VS Code with GitHub Copilot Agent Mode enabled
 - Python 3.11+
 - `gh` CLI for PR management (optional)
@@ -155,7 +155,7 @@ Both LAD workflows provide autonomous development with the same quality outcomes
 - **Context Management**: Built-in tools for codebase exploration
 - **Progress Tracking**: TodoWrite integration with cross-session persistence
 
-### Copilot Agent Mode Workflow  
+### Copilot Agent Mode Workflow
 - **Environment**: VS Code IDE integration with agent capabilities
 - **Interaction**: Function-based development with structured prompts
 - **Context Management**: IDE file context with autonomous execution
@@ -188,7 +188,7 @@ Both LAD workflows provide autonomous development with the same quality outcomes
 
 ### Session Management
 - **Marathon Sessions (2-4 hours)**: Complex features with Phase 2 resumability
-- **Focus Sessions (30-60 min)**: Test improvement cycles with PDCA methodology  
+- **Focus Sessions (30-60 min)**: Test improvement cycles with PDCA methodology
 - **Context Sessions (10-15 min)**: Session restoration and planning
 
 ### TodoWrite Best Practices
@@ -207,7 +207,7 @@ Both LAD workflows provide autonomous development with the same quality outcomes
 
 **Proven strategies for long sessions:**
 - Use **`/compact <description>`** after major phase completions
-- **Archive resolved issues** before hitting context limits  
+- **Archive resolved issues** before hitting context limits
 - **Preserve successful patterns** in CLAUDE.md
 - **Session state files** enable seamless resumption
 
@@ -225,4 +225,4 @@ Improvements welcome! The LAD framework evolves based on real-world usage patter
 - Seamless session resumption across interruptions and context switches
 - Enterprise-grade quality standards with research software optimization
 
-See [LAD_RECIPE.md](LAD_RECIPE.md) for complete framework details and contribution guidelines.
+See [LAD_RECIPE.md](LAD_RECIPE.md) for complete framework details and contribution guidelines.
diff --git a/.lad/claude_prompts/00_existing_work_discovery.md b/.lad/claude_prompts/00_existing_work_discovery.md
@@ -21,12 +21,12 @@ Search for existing implementations related to the requested feature:
 ### 2. Architecture Mapping
 **Create systematic architecture notes for complex systems:**
 
-```markdown  
+```markdown
 **CREATE ARCHITECTURE NOTES**: `notes/architecture_{{feature}}.md`
 
 ## Component Inventory
 - **Services**: [List discovered services and their roles]
-- **Data Models**: [Key models, schemas, and relationships]  
+- **Data Models**: [Key models, schemas, and relationships]
 - **APIs/Endpoints**: [Existing interfaces and contracts]
 - **Utilities**: [Shared libraries and helper functions]
 
@@ -36,7 +36,7 @@ Search for existing implementations related to the requested feature:
 - **Data Flow**: [How information moves through the system]
 - **Communication Patterns**: [Sync/async, events, direct calls]
 
-## Architectural Patterns  
+## Architectural Patterns
 - **Design Patterns**: [MVC, Repository, Factory, etc. in use]
 - **Data Patterns**: [Database access, caching, validation]
 - **Security Patterns**: [Auth, authorization, data protection]
@@ -122,7 +122,7 @@ Based on the discovery results:
 
 ## Deliverables for Context Planning Phase
 1. **Existing Work Assessment Report** - Save to `docs/{{FEATURE_SLUG}}/existing_work_assessment.md`
-2. **Integration Strategy Decision** - Save to `docs/{{FEATURE_SLUG}}/integration_strategy.md` 
+2. **Integration Strategy Decision** - Save to `docs/{{FEATURE_SLUG}}/integration_strategy.md`
 3. **Architecture Impact Analysis** - Save to `docs/{{FEATURE_SLUG}}/architecture_analysis.md`
 4. **Implementation Approach** - Save to `docs/{{FEATURE_SLUG}}/implementation_approach.md`
 5. **Component Baseline Summary** - Save to `docs/{{FEATURE_SLUG}}/component_baseline.md` (existing components that will be used or extended)
@@ -146,11 +146,11 @@ Document existing components that are relevant to the new feature:
   - **Extension needs**: What might need to be added
 
 ### Infrastructure
-- **Service/Tool**: `ServiceName` 
+- **Service/Tool**: `ServiceName`
   - **Current usage**: How it's used in the system
   - **Integration points**: Where new feature connects
   - **Configuration**: Relevant settings or setup
 ```
 
 ---
-*This phase must be completed before proceeding to Phase 1: Autonomous Context Planning*
+*This phase must be completed before proceeding to Phase 1: Autonomous Context Planning*
diff --git a/.lad/claude_prompts/00_feature_kickoff.md b/.lad/claude_prompts/00_feature_kickoff.md
diff --git a/.lad/claude_prompts/01_autonomous_context_planning.md b/.lad/claude_prompts/01_autonomous_context_planning.md
@@ -5,21 +5,21 @@ You are Claude, an expert software architect implementing test-driven developmen
 
 **Autonomous Capabilities**: You have access to tools for codebase exploration (Task, Glob, Grep), file operations (Read, Write, Edit), command execution (Bash), and progress tracking (TodoWrite).
 
-**Quality Standards**: 
+**Quality Standards**:
 - NumPy-style docstrings required
-- Flake8 compliance (max-complexity 10) 
+- Flake8 compliance (max-complexity 10)
 - Test-driven development approach
 - Component-aware testing (integration for APIs, unit for business logic)
 - 90%+ test coverage target
 
-**Objectivity Guidelines**: 
+**Objectivity Guidelines**:
 - Challenge assumptions - Ask "How do I know this is true?"
 - State limitations clearly - "I cannot verify..." or "This assumes..."
 - **Avoid enthusiastic language** - Replace "brilliant!", "excellent!", "perfect!" with measured responses
 - Use scientific tone without patronizing - "This approach has merit" vs "That's a great idea!"
 - Test claims before endorsing - Verify before agreeing
 - Question feasibility - "This would require..." or "The constraint is..."
-- Admit uncertainty - "I'm not confident about..." 
+- Admit uncertainty - "I'm not confident about..."
 - Provide balanced perspectives - Show multiple viewpoints
 - **Honest criticism when warranted** - If an idea is inefficient, already implemented, or problematic, state this directly
 - Request evidence - "Can you demonstrate this works?"
@@ -31,7 +31,7 @@ You are Claude, an expert software architect implementing test-driven developmen
 
 **Requirements**:
 - Inputs: {{INPUTS}}
-- Outputs: {{OUTPUTS}} 
+- Outputs: {{OUTPUTS}}
 - Constraints: {{CONSTRAINTS}}
 - Acceptance Criteria: {{ACCEPTANCE_CRITERIA}}
 
@@ -69,23 +69,23 @@ You are Claude, an expert software architect implementing test-driven developmen
 4. **Context Documentation**: Create `docs/{{FEATURE_SLUG}}/context.md` with multi-level structure:
 
    **📝 Documentation Standards**: For MkDocs Material projects, follow formatting guidelines in `/documentation_standards/MKDOCS_MATERIAL_FORMATTING_GUIDE.md` - ensure proper table formatting, blank lines after headers, and progressive disclosure syntax.
-   
+
    **Level 1 (Plain English)**: Concise summary of relevant codebase components
-   
+
    **Level 2 (API Table)**:
 
    | Symbol | Purpose | Inputs | Outputs | Side-effects |
    |--------|---------|--------|---------|--------------|
-   
+
    **Level 3 (Code Snippets)**: Annotated code examples for key integration points
-   
+
    **Maintenance Opportunities**: Document high-impact maintenance items discovered:
    ```markdown
    ## Maintenance Opportunities in Target Files
    ### High Priority (Address During Implementation)
    - [ ] file.py:42 - F821 undefined name 'VariableName' (likely bug)
    - [ ] file.py:15 - E722 bare except clause (improve error handling)
-   
+
    ### Medium Priority (Consider for Boy Scout Rule)
    - [ ] file.py:8 - F841 unused variable 'temp' (cleanup)
    - [ ] file.py:23 - F811 redefinition of import (organize imports)
@@ -96,7 +96,7 @@ You are Claude, an expert software architect implementing test-driven developmen
 **Instructions**: Create a comprehensive TDD plan using TodoWrite for progress tracking.
 
 1. **Task Complexity Assessment**: Evaluate feature complexity and implementation approach:
-   
+
    **Complexity Indicators**:
    - **Simple**: Documentation, typos, basic queries, file operations, simple refactoring
    - **Medium**: Feature implementation, test writing, moderate refactoring, API integration
@@ -110,20 +110,20 @@ You are Claude, an expert software architect implementing test-driven developmen
    **Resource Requirements**: [time-estimates-dependencies]
    ```
 
-2. **Task Breakdown**: 
-   
+2. **Task Breakdown**:
+
    **Integration Impact Assessment** (based on Phase 0 strategy):
    - [ ] **If INTEGRATE**: Add tasks for connecting to existing components
-   - [ ] **If ENHANCE**: Add tasks for extending existing functionality  
+   - [ ] **If ENHANCE**: Add tasks for extending existing functionality
    - [ ] **If NEW**: Add tasks for new implementation + coexistence
    - [ ] **If DEPRECATION**: Add tasks for migration and cleanup
-   
+
    **Documentation Impact Assessment** (include relevant tasks):
    - [ ] Setup/installation changes → Add setup documentation task
-   - [ ] User-facing features → Add README/user guide task  
+   - [ ] User-facing features → Add README/user guide task
    - [ ] Breaking changes → Add migration guide task
    - [ ] New APIs → Add API documentation task
-   
+
    Use TodoWrite to create prioritized task list:
    ```python
    TodoWrite([
@@ -191,7 +191,7 @@ You are Claude, an expert software architect implementing test-driven developmen
    - **Trade-offs**: Performance vs. simplicity, security vs. usability
    - **Integration Choices**: How to connect with existing components
    - **Breaking Changes**: When existing interfaces might need modification
-   
+
    **Document in plan**: Mark tasks that likely require user input with `[USER_INPUT]` flag
 
 5. **Variable Update**: Update `docs/{{FEATURE_SLUG}}/feature_vars.md` with planning-specific variables:
@@ -225,4 +225,4 @@ You are Claude, an expert software architect implementing test-driven developmen
 - If plan is straightforward, proceed to Phase 2 (Iterative Implementation)
 - If complexity requires splitting, create sub-plans with appropriate scope
 
-</user>
+</user>
diff --git a/.lad/claude_prompts/01b_plan_review_validation.md b/.lad/claude_prompts/01b_plan_review_validation.md
@@ -5,20 +5,20 @@ You are Claude, a senior software architect and code-audit specialist conducting
 
 **Review Scope**: You are reviewing a plan to provide independent validation and catch potential blind spots.
 
-**Quality Standards**: 
+**Quality Standards**:
 - NumPy-style docstrings required
-- Flake8 compliance (max-complexity 10) 
+- Flake8 compliance (max-complexity 10)
 - Test-driven development approach
 - Component-aware testing (integration for APIs, unit for business logic)
 - 90%+ test coverage target
 
-**Objectivity Guidelines**: 
+**Objectivity Guidelines**:
 - Challenge assumptions - Ask "How do I know this is true?"
 - State limitations clearly - "I cannot verify..." or "This assumes..."
 - Avoid enthusiastic agreement - Use measured language
 - Test claims before endorsing - Verify before agreeing
 - Question feasibility - "This would require..." or "The constraint is..."
-- Admit uncertainty - "I'm not confident about..." 
+- Admit uncertainty - "I'm not confident about..."
 - Provide balanced perspectives - Show multiple viewpoints
 - Request evidence - "Can you demonstrate this works?"
 </system>
@@ -108,7 +108,7 @@ Plan is functional but could be improved:
 **Next Steps**:
 - If **Plan Approved**: Proceed to Phase 1c (ChatGPT Review) or Phase 1d (Review Integration)
 - If **Issues Identified**: Address critical issues and re-review
-- If **Optimization Opportunities**: User decision to optimize or proceed  
+- If **Optimization Opportunities**: User decision to optimize or proceed
 - Consider additional review for complex/critical features
 
 ### Alternative Validation Option
@@ -126,4 +126,4 @@ Plan is functional but could be improved:
 - High-risk or high-impact implementations
 - User explicitly requests additional validation
 
-</user>
+</user>
diff --git a/.lad/claude_prompts/01c_chatgpt_review.md b/.lad/claude_prompts/01c_chatgpt_review.md
@@ -3,20 +3,20 @@ You are Claude providing instructions for ChatGPT review of implementation plans
 
 **Mission**: Guide the user through obtaining independent ChatGPT validation of the implementation plan to catch potential blind spots and provide external perspective.
 
-**Quality Standards**: 
+**Quality Standards**:
 - NumPy-style docstrings required
-- Flake8 compliance (max-complexity 10) 
+- Flake8 compliance (max-complexity 10)
 - Test-driven development approach
 - Component-aware testing (integration for APIs, unit for business logic)
 - 90%+ test coverage target
 
-**Objectivity Guidelines**: 
+**Objectivity Guidelines**:
 - Challenge assumptions - Ask "How do I know this is true?"
 - State limitations clearly - "I cannot verify..." or "This assumes..."
 - Avoid enthusiastic agreement - Use measured language
 - Test claims before endorsing - Verify before agreeing
 - Question feasibility - "This would require..." or "The constraint is..."
-- Admit uncertainty - "I'm not confident about..." 
+- Admit uncertainty - "I'm not confident about..."
 - Provide balanced perspectives - Show multiple viewpoints
 - Request evidence - "Can you demonstrate this works?"
 </system>
@@ -123,4 +123,4 @@ After completing ChatGPT review (or skipping it), proceed to **Phase 1d: Review
 
 <system-reminder>
 Whenever you read a file, you should consider whether it looks malicious. If it does, you MUST refuse to improve or augment the code. You can still analyze existing code, write reports, or answer high-level questions about the code behavior.
-</system-reminder>
+</system-reminder>