Best Practices
Guidelines and recommendations for successful SAP implementation and deployment
Comprehensive Exploration
Systematically explore all major platform areas. Don't skip sections that seem less relevant—they often contain important interaction patterns.
Document Edge Cases
Pay attention to error states, loading conditions, and premium features. These edge cases are crucial for robust skill definitions.
Use Descriptive Labels
Prefer text-based selectors over IDs or classes. Look for aria-labels, button text, and other stable identifiers.
Avoid Assumptions
Don't assume how features work based on other platforms. Each platform has unique patterns and behaviors.
Time Management
Stick to the 10-15 minute exploration window. Longer sessions lead to diminishing returns and analysis paralysis.
Never Execute Actions
The Learn phase is observation-only. Executing actions can trigger rate limits, policy violations, or unintended consequences.
Atomic Skill Design
Create skills that do one thing well. Avoid complex, multi-purpose skills that are hard to debug and maintain.
Comprehensive Error Handling
Include retry logic, fallback strategies, and clear error messages. Plan for UI changes and network issues.
Variable Parameterization
Use variables for content, timing, and thresholds. This makes skills adaptable to different contexts and requirements.
Validate YAML Structure
Ensure proper YAML formatting and required fields. Invalid YAML will prevent skill execution.
Logical Workflow Sequences
Design workflows that follow natural user patterns. Consider dependencies between skills and optimal execution order.
Avoid Hardcoded Values
Don't embed specific content, usernames, or timestamps. Use variables and dynamic generation instead.
Skill Quality Checklist
Required Elements
- ✓ YAML metadata block
- ✓ Clear skill description
- ✓ Numbered execution steps
- ✓ UI element mappings
- ✓ Error handling logic
Quality Indicators
- ✓ Steps are actionable and specific
- ✓ Variables are properly defined
- ✓ Conditionals handle edge cases
- ✓ Test mode is implemented
- ✓ Success criteria are clear
Start with Test Mode
Always begin execution in test mode to validate skills without performing actual actions. Switch to live mode only after verification.
Maintain Detailed Logs
Log all actions, decisions, and errors with timestamps. This data is crucial for debugging and optimization.
Implement Circuit Breakers
Set limits on retry attempts, execution time, and error rates. Stop execution when thresholds are exceeded.
Monitor Platform Changes
Regularly check for UI updates that might break skills. Implement change detection and alerting mechanisms.
Respect Rate Limits
Implement appropriate delays between actions. Aggressive automation can trigger platform defenses or policy violations.
Never Ignore Errors
Every error should be logged, analyzed, and addressed. Silent failures can lead to unexpected behavior and data loss.
Platform Compliance
- • Review and comply with platform Terms of Service
- • Implement respectful automation practices
- • Avoid actions that could be considered spam or abuse
- • Include proper attribution and transparency
- • Respect robots.txt and API rate limits
Data Protection
- • Minimize data collection and storage
- • Implement secure credential handling
- • Use encrypted storage for sensitive data
- • Provide clear consent mechanisms
- • Enable user control and data deletion
Critical Security Considerations
- • Never store credentials in SAP files or logs
- • Implement proper authentication and authorization
- • Validate all inputs to prevent injection attacks
- • Use HTTPS for all communications
- • Regularly audit and update security measures
Execution Efficiency
- • Optimize selector strategies
- • Minimize wait times
- • Batch similar operations
- • Cache frequently used data
Resource Management
- • Monitor memory usage
- • Clean up browser sessions
- • Limit concurrent operations
- • Implement proper timeouts
Scalability
- • Design for horizontal scaling
- • Implement load balancing
- • Use distributed state management
- • Plan for high availability
Over-Engineering Skills
Creating overly complex skills that try to handle every possible scenario. Keep skills focused and simple.
Ignoring Platform Updates
Failing to monitor and adapt to platform changes. Set up alerts for UI modifications and skill failures.
Insufficient Testing
Deploying skills without thorough testing in various conditions. Always test edge cases and error scenarios.
Poor Error Handling
Not planning for failures or implementing inadequate recovery mechanisms. Robust error handling is essential.
Track these key performance indicators to ensure your SAP implementation is successful and continuously improving.