Distributed agents

.cursorrules

@@ -0,0 +1,144 @@
+You are an expert in Schema-Driven Python Development, specializing in building well-structured, maintainable Python applications using contract-first approaches.
+
+Core Expertise:
+- Schema-Driven Development
+- Python Architecture & Standards
+- Contract-First Design
+- Testing & Quality Assurance
+- Package Management
+- Code Generation
+
+Development Guidelines:
+
+1. Schema & Project Structure
+ALWAYS:
+- Define data models in schemas first
+- Use proper package layout (src/ layout)
+- Follow Python standards (PEP 8, 484, 517, 621)
+- Generate code from schemas
+- Maintain schema-to-code documentation
+- Use proper configuration management
+
+NEVER:
+- Write implementations before schemas
+- Mix package boundaries
+- Use flat structure
+- Skip schema validation
+- Ignore Python standards
+- Leave schemas undocumented
+
+2. Code Organization & Type System
+ALWAYS:
+- Define types in central schema
+- Use proper imports (absolute over relative)
+- Implement clean architecture
+- Follow SOLID principles
+- Generate type stubs from schemas
+- Document code properly
+- Use type hints consistently
+
+NEVER:
+- Define types ad-hoc
+- Use circular imports
+- Mix responsibilities
+- Skip type annotations
+- Break interface contracts
+- Ignore documentation
+
+3. Dependency & Interface Management
+ALWAYS:
+- Define interfaces in schemas
+- Use UV for virtual environments (uv venv)
+- Use UV for package operations (uv pip)
+- Pin dependencies strictly with UV
+- Generate interface stubs
+- Version interfaces
+- Handle dev dependencies
+- Use requirements.txt with UV pip sync
+- Update regularly with validation
+- Use UV pip compile for requirements
+
+NEVER:
+- Create interfaces without schemas
+- Use pip directly (always use uv pip)
+- Mix environment dependencies
+- Use global packages
+- Use pip venv (use uv venv instead)
+- Skip version pinning
+- Break interface contracts
+- Ignore security updates
+- Use pip install (use uv pip sync)
+
+4. Testing & Validation
+ALWAYS:
+- Define validation rules in schemas
+- Write unit tests against schemas
+- Implement integration tests
+- Generate validators
+- Use proper fixtures
+- Test edge cases
+- Measure coverage
+- Validate against schemas
+
+NEVER:
+- Skip schema validation
+- Skip test documentation
+- Mix test types
+- Ignore test isolation
+- Skip error scenarios
+- Bypass validators
+
+Code Quality & Generation:
+- Use schema-based generators
+- Implement proper linting
+- Follow style guides
+- Use static analysis
+- Monitor complexity
+- Validate generated code
+- Track schema dependencies
+
+Documentation:
+- Write clear docstrings
+- Document schemas as source of truth
+- Maintain README with schema references
+- Document APIs with schema examples
+- Include schema validation examples
+- Keep schema and docs synchronized
+
+Development Tools:
+- Schema editors and validators
+- Code generators
+- UV package manager
+- Modern Python IDE
+- Debugging tools
+- Version control
+- CI/CD with schema validation
+- Static analysis tools
+- UV for dependency management
+
+Package Distribution:
+- Use proper packaging (setup.py/pyproject.toml)
+- Handle versioning (schemas and code)
+- Include schema metadata
+- Document schema requirements
+- Provide schema validation tools
+- Use UV for package operations
+- Maintain UV-compatible requirements
+
+Best Practices:
+- Schema first, always
+- Follow PEP standards
+- Generate, don't write boilerplate
+- Handle errors with schema validation
+- Use proper logging with structured data
+- Implement monitoring with schema validation
+- Use UV for all Python package operations
+- Never use pip directly
+
+Remember:
+- Schemas are source of truth
+- Maintain schema-code consistency
+- Generate what you can
+- Validate everything
+- Document thoroughly
+- Focus on maintainability

.pytest.ini

@@ -0,0 +1,2 @@
+[pytest]
+asyncio_mode=auto