Predictive Code Analysis I'll analyze your codebase to predict potential problems before they impact your project. ## Strategic Thinking Process <think> To make accurate predictions, I need to consider: 1. Pattern Recognition - Which code patterns commonly lead to problems? - Are there areas of increasing complexity? - Do I see anti-patterns that will cause issues at scale? - Are there ticking time bombs (hardcoded values, assumptions)? 2. Risk Assessment Framework - Likelihood: How likely is this issue to occur? - Impact: How severe would the consequences be? - Timeline: When might this become a problem? - Effort: How difficult would it be to fix now versus later? 3. Common Problem Categories - Performance: O(n²) algorithms, memory leaks, inefficient queries - Maintainability: High complexity, poor naming, tight coupling - Security: Input validation gaps, exposed secrets, weak authentication - Scalability: Hardcoded limits, single points of failure 4. Prediction Strategy - Start with the highest-risk areas (critical path code) - Look for patterns that break at 10x or 100x scale - Check for the accumulation of technical debt - Identify fragility at integration points </think> Based on this analysis framework, I’ll use native tools for comprehensive analysis: - Grep tool to search for problematic patterns - Glob tool to analyze file structures and growth - Read tool to examine complex functions and hotspots I’ll examine: - Trends in code complexity and potential hotspots - Emerging performance bottleneck patterns - Indicators of maintenance difficulty - Architectural stress points and scaling issues - Gaps in error handling For each prediction, I’ll: - Highlight specific code locations with file references - Explain why it’s likely to cause future issues - Estimate the potential timeline and impact - Suggest preventive measures with priority levels When I find multiple issues, I’ll create a to-do list for systematic review and prioritization. Areas of analysis: - Functions approaching complexity thresholds - Files with high change frequency (potential hotspots) - Dependencies with known issues or update requirements - Performance patterns that do not scale - Code duplication leading to maintenance issues After the analysis, I’ll ask: “How would you like to track these predictions?” - Create to-dos: I’ll add items to track resolution progress - Create issues on GitHub: I’ll generate properly formatted issues with details - Summary only: I’ll provide an actionable report without creating tasks Important: I will NEVER: - Add “Created by Claude” or any AI attribution to issues - Include “Generated with Claude Code” in descriptions - Modify repository settings or permissions - Add any AI/assistant signatures or watermarks Predictions will include: - Risk level assessment (Critical/High/Medium/Low) - Estimated timeline for potential issues - Specific remediation recommendations - Impact assessment on project goals This helps prevent problems before they impact your project, saving time and proactively maintaining code quality.