Structured Workflow Engine MCP Server

# Code Analysis Prompt ## 🎯 Goal Deep-scan the codebase, expose issues, and recommend fixes—no fluff. ## 📥 Context (ask if missing) 1. **Scope** – specific modules/files to inspect. 2. **Repo Access** – git URL or path. 3. **Known Issues** – bugs, perf pains, security worries. 4. **Focus** – pick the priority: architecture / performance / security / maintainability. ## 🚦 Skip if - The code is trivial **and** unchanged, or a fresh analysis exists (<30 days). ## 🔍 Checklist - **Structure** - [ ] File & module layout - [ ] Architecture patterns - [ ] Dependency graph & coupling - [ ] Interfaces / public APIs - **Quality** - [ ] Readability (naming, comments, style) - [ ] Maintainability & extensibility - [ ] Complexity hotspots (length, nesting, cyclomatic) - [ ] Code smells & duplication - **Perf / Sec** - [ ] Algorithmic complexity on critical paths - [ ] Resource usage (memory, DB, I/O) - [ ] Security: input validation, authN/Z, crypto, secrets - [ ] Error & exception handling patterns ## 📤 Outputs (in `docs/analysis/[date]-report.md`), which includes: - overview + dependency diagram (Mermaid/PlantUML) - metrics, smells, maintainability score - prioritized table with file/line refs - quick wins, tech-debt roadmap, effort estimates ## ➡️ Response Flow ```mermaid flowchart LR U[User] -->|request| A[Code Analysis Engine] A --> B{Need input?} B -- Yes --> C[Ask for repo / scope] B -- No --> D[Run analysis & generate reports]