# Reverse Engineering > This skill comprehensively analyzes existing codebases and extracts specifications, feature documentation, and technical documentation to serve as a foundation for modernization efforts. Use when this capability is needed. - Skill: `tomevault-io/reverse-engineering-3` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/reverse-engineering-3` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/reverse-engineering-3/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Docs & Writing - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/reverse-engineering-3 --- --- name: reverse-engineering description: Analyzes existing codebases to extract specifications, create feature documentation, and generate comprehensive technical documentation. Use this skill when asked to analyze a codebase, document existing code, reverse engineer an application, extract specifications, or create technical documentation from code. --- # Reverse Engineering Skill This skill comprehensively analyzes existing codebases and extracts specifications, feature documentation, and technical documentation to serve as a foundation for modernization efforts. ## When to Use This Skill - Analyzing an existing codebase - Extracting specifications from legacy code - Creating technical documentation from code - Preparing for modernization or migration - Understanding undocumented systems ## Critical Rules ### What You Must NEVER Do 1. **NEVER modify, create, or change ANY code** - Documentation ONLY 2. **NEVER make things up** - Document ONLY what actually exists 3. **NEVER create `specs/prd.md`** - That is PM's responsibility 4. **NEVER fabricate details** - If it doesn't exist, say so 5. **NEVER assume missing implementations** - Be honest about gaps ### What You Must Do 1. **Document what EXISTS** - Analyze the actual implementation as-is 2. **Be honest about gaps** - Explicitly state missing or incomplete elements 3. **Map to real code** - Every finding must reference actual files 4. **Accept any tech stack** - Work with whatever the project uses 5. **Be comprehensive** - Leave no stone unturned ## Analysis Workflow ### Phase 1: Discovery and Inventory #### 1. Repository Discovery - Read configuration files (package.json, requirements.txt, .csproj, etc.) - Examine project structure and directory purposes - Identify entry points and startup configurations - Catalog build and deployment configurations #### 2. Technology Stack Mapping - Detect all programming languages used - Identify frameworks, ORMs, testing frameworks - Inventory build tools and package managers - Capture versions where possible ### Phase 2: Feature Extraction #### 3. Feature Discovery - Identify all user-facing features and pages - Document all API endpoints and functionality - Extract core business processes - Examine database schemas and data structures #### 4. Business Feature Analysis Create individual files in `specs/features/` with: - Feature purpose and business problem solved - User workflows and interactions - Functional requirements - Acceptance criteria (from observable behavior) - Dependencies on other features - Implementation status (complete/partial/missing) ### Phase 3: Technical Deep Dive #### 5. Architecture Assessment - Map component relationships and interactions - Trace data flow through the system - Identify external system connections - Document architectural and design patterns #### 6. Quality Assessment - Evaluate existing test coverage - Assess documentation completeness - Identify maintainability issues - Document security patterns and concerns ## Documentation Structure to Create ``` specs/ ├── features/ # Feature Requirements Documents │ ├── [feature-1].md │ ├── [feature-2].md │ └── ... └── docs/ ├── architecture/ │ ├── overview.md # System architecture overview │ ├── components.md # Component relationships │ ├── patterns.md # Design patterns used │ └── security.md # Security patterns ├── technology/ │ ├── stack.md # Complete technology inventory │ ├── dependencies.md # All dependencies with versions │ └── tools.md # Build and dev tools ├── infrastructure/ │ ├── deployment.md # Deployment architecture │ ├── environments.md # Environment configurations │ └── operations.md # Operational procedures └── integration/ ├── apis.md # API documentation ├── databases.md # Database schemas └── services.md # External services ``` ## Success Criteria Analysis is complete when: - ✅ Complete documentation structure created - ✅ All major features identified and documented - ✅ Technology stack completely mapped - ✅ Architecture clearly documented - ✅ Dependencies catalogued - ✅ Integration points mapped - ✅ Code quality assessed honestly - ✅ Everything linked to actual code files ## Next Steps After analysis is complete: - Hand off to **Modernizer Agent** for modernization planning - Feature documentation guides functionality preservation - Technical assessment informs priorities and risk management ## Templates See `templates/` for documentation templates. ## Sample Output See `examples/sample-analysis-output.md` for a complete example. --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/henrybravo) — claim your Tome and manage your conversions.