Production-ready modular Claude Code framework with 30+ commands, token optimization, and MCP server integration. Achieves 2-10x productivity gains through systematic command organization and hierarchical configuration.
git clone https://github.com/oxygen-fragment/claude-modular.gitClaude Modular is a comprehensive framework template for Claude Code that organizes development workflows into modular, reusable commands across project management, development, testing, deployment, and documentation. The framework achieves 50-80% token savings through progressive disclosure and context compression, enabling 2-10x productivity gains for teams building with Claude. It provides 20+ production-ready commands following proven XML structures, layered configuration management for different environments, and built-in security controls including secret scanning, permission validation, and audit logging. Teams can customize the framework for their tech stack while maintaining consistent code review quality, automated testing coverage, and standardized deployment procedures. Integration with MCP servers extends functionality to version control, file operations, and issue tracking systems.
[{"step":"Install claude-modular framework","action":"Run `pip install claude-modular` or clone from [GitHub repository]. Configure MCP servers in `config/mcp_servers.json` for your workflow.","tip":"Enable token optimization in settings.json with 'token_optimizer': {'enabled': true, 'chunk_size': 2000}"},{"step":"Select appropriate modules for your task","action":"Use `modular list` to view available commands. For data processing, use `modular system_analyzer`. For code generation, use `modular code_generator`.","tip":"Combine modules hierarchically: `modular system_analyzer | modular service_designer | modular code_generator`"},{"step":"Configure hierarchical configuration","action":"Create a `config/project.json` file with task-specific settings. Example: {'project': 'api_migration', 'services': ['user', 'order', 'payment'], 'dependencies': ['redis', 'postgres']}","tip":"Use `modular config init` to generate template configuration files"},{"step":"Execute and validate","action":"Run `modular execute [TASK] --config config/project.json` and review the generated artifacts. Use `modular validate` to check for errors.","tip":"Enable verbose logging with `--log-level debug` for troubleshooting complex workflows"},{"step":"Iterate and optimize","action":"Use `modular analyze` to review token usage and performance metrics. Adjust chunk sizes and MCP integrations based on findings.","tip":"Save optimized configurations as templates in `config/templates/` for reuse across projects"}]
Setting up new Claude Code projects with standardized command libraries and environment configuration
Automating code review, testing, and documentation generation across development teams
Managing feature creation and component scaffolding with consistent boilerplate and quality gates
Deploying to staging and production with validation checkpoints and rollback procedures
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/oxygen-fragment/claude-modularCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Use the claude-modular framework to [TASK]. Structure your approach using the modular command system: [SPECIFIC_MODULE] for [SUBTASK_1], [ANOTHER_MODULE] for [SUBTASK_2], and [FINAL_MODULE] for [SUBTASK_3]. Optimize token usage by [OPTIMIZATION_STRATEGY]. Integrate MCP servers where relevant to [INTEGRATION_PURPOSE]. Document each step in the [OUTPUT_FORMAT].
``` # claude-modular Framework Execution Report ## Project: E-commerce API Migration ### Task: Migrate legacy monolith to microservices ### Phase 1: System Analysis (modular/system_analyzer) - **Token Usage**: 1,247 tokens (optimized via chunked analysis) - **Findings**: - 4 critical monolith modules identified - 12 shared dependencies with tight coupling - API endpoint count: 87 (legacy) → 23 (target microservices) - **MCP Integration**: Used `mcp-server-git` to analyze 2,412 commits for dependency patterns ### Phase 2: Service Design (modular/service_designer) - **Commands Executed**: 1. `service_boundary_analyzer` → Identified 3 bounded contexts 2. `dependency_graph_visualizer` → Generated DOT file for Graphviz 3. `api_endpoint_mapper` → Created migration matrix (87→23 endpoints) - **Output**: service_design.json with 15 service definitions ### Phase 3: Migration Scripting (modular/code_generator) - **Generated Artifacts**: - 23 Dockerfiles (one per service) - 15 Terraform modules for infrastructure - 87 migration scripts (one per endpoint) - **Token Savings**: 40% reduction via template reuse ### Phase 4: Validation (modular/test_automator) - **Test Coverage**: 94% (legacy: 68%) - **Performance Metrics**: - Cold start: 1.2s (target: <2s) - Memory usage: 45MB/service (target: <50MB) - **MCP Integration**: `mcp-server-kubernetes` for cluster validation ## Recommendations 1. Prioritize UserService migration (highest business impact) 2. Implement circuit breakers for PaymentService dependencies 3. Schedule blue-green deployment for OrderService (highest risk) ```
Control SaaS spending with visibility and analytics
AI assistant built for thoughtful, nuanced conversation
Auto-transcribe meetings and generate action items
IronCalc is a spreadsheet engine and ecosystem
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan