chore: add Cursor/Claude Code configuration

Added based on Feb 2026 Cursor New Features Session recommendations:

- .cursorrules: Simple one-liner rules (GPG signing, commit format, testing)
- .cursorignore: Exclude large files from AI context (improves performance)
- .claude/: Folder structure for AI working files (gitignored)
  - README.md: Documentation for folder structure
  - context/cursor-claude-best-practices.md: Best practices from meetings

Key improvements:
- Better AI context management (excludes models, screenshots, lock files)
- Clear separation: rules (simple) vs skills (complex workflows)
- PM-thinking approach from Amit meeting integrated
- Follows Cursor's rules → skills migration path

Related: W-21315032

Author: alexpmule

.cursorignore

@@ -0,0 +1,47 @@
+# Cursor Ignore - Exclude from AI context for better performance
+
+# Build outputs
+dist/
+coverage/
+demo/vendor.js
+
+# Dependencies
+node_modules/
+
+# Generated AMF models (except specific test models)
+demo/models/*.json
+demo/models/flattened/*.json
+!demo/models/jldAsync26.json
+!demo/models/avro.json
+!demo/models/avro2.json
+!demo/models/grpc-test.json
+!demo/models/grpc-test-compact.json
+!demo/models/product-order-deep-allof.json
+
+# IDE configs (keep lightweight)
+.vscode/*
+!.vscode/settings.json
+.idea/
+
+# OS files
+.DS_Store
+Thumbs.db
+*.stackdump
+
+# Git internals
+.git/
+
+# Large log files
+*.log
+npm-debug.log*
+
+# Test screenshots (visual regression baselines - too many files)
+test/visual/screenshots/
+
+# Package lock files (too large, let Cursor use package.json)
+package-lock.json
+
+# Temporary files
+*.tmp
+*.swp
+*~

.cursorrules

@@ -0,0 +1,61 @@
+# Cursor Rules for api-console v6
+
+## Repository Context
+- This is api-console v6, built with LitElement and Web Components
+- Open source project (CPAL-1.0 license)
+- Part of MuleSoft's API Console ecosystem
+
+## Git & Commit Rules
+- NEVER commit directly to master/main (ALWAYS create feature branch first)
+- ALWAYS run `mulesoft-git` before committing (GPG sign with alexperez@mulesoft.com)
+- Use Conventional Commits format: `<type>(<scope>): <subject>`
+- Valid types: feat, fix, docs, style, refactor, test, chore, perf, ci, build, revert
+
+## Branch Naming Conventions
+- Component fixes: `fix/W-XXXXXXXX-description`
+- Features: `feat/W-XXXXXXXX-description` or `feat/description`
+- Releases: `build/X.X.X` (api-console only)
+- Chores: `chore/description`
+
+## Code Style
+- Use @web/test-runner for testing (NOT Jest - poor Shadow DOM support)
+- Use AmfHelperMixin utilities for AMF queries (NOT direct JSON access)
+- Custom events for component communication (NOT direct method calls)
+- Shadow DOM with CSS custom properties for theming
+- No `any` types in TypeScript
+- Prefer Edit tool over Write for existing files
+
+## Testing
+- Write tests immediately after implementing feature (TDD approach)
+- Run `npm test` before creating PR (must pass)
+- Update visual regression baselines when UI changes intentionally: `npm run test:visual:update`
+
+## Build & Verification
+- DO NOT run build after every small change (saves tokens, avoids loops)
+- Order: Implement → Tests → Build → Fix compilation errors
+- Only build when: user asks explicitly, before PR, or after major type changes
+
+## AMF Model Architecture
+- API Console does NOT parse API specs directly (requires AMF model)
+- Models are JSON-LD format (AMF v4.0+)
+- Generate models: `npm run build:models`
+- Load via `amf` property or `modelLocation` (URL to JSON)
+
+## Common Pitfalls
+- Don't parse API files directly (use AMF models)
+- Don't import CodeMirror/crypto as ES modules (use vendor.js)
+- Don't expose Entities in API responses (use DTOs)
+- Don't put business logic in Controllers (use Service layer)
+- Don't modify AMF model structure (external standard)
+
+## Quick Commands
+- `npm run prepare` - Build vendor.js + generate AMF models (after clone)
+- `npm start` - Dev server
+- `npm test` - Run all tests
+- `npm run build:vendor` - Rebuild non-ES6 dependencies
+
+## Related Documentation
+- CLAUDE.md - Full project context
+- docs/team/patterns/ - Architectural decisions
+- docs/team/runbooks/ - Operational procedures
+- docs/team/onboarding/ - New team member guide