# HUMMBL Claude Skills Production-ready Claude skills for HUMMBL mental models framework development, multi-agent coordination, and repository management. [](LICENSE) []() []() ## Overview This repository contains 6 production-ready Claude skills following the [Anthropic Agent Skills](https://docs.claude.com/en/docs/agents-and-tools/agent-skills) specification. Each skill provides comprehensive guidance, templates, and examples for specific domains. **Total Lines of Expertise:** 3,359 lines of production-grade guidance ## Skills Catalog ### 1. MCP Server Developer **File:** \`mcp-server-developer/\` **Lines:** 425 **Purpose:** Model Context Protocol (MCP) server implementation for Claude Desktop **Capabilities:** - Complete MCP protocol implementation guidance - TypeScript + Node.js patterns - D1 SQLite + FTS5 integration - Telemetry and analytics setup - npx package distribution - Phase 0-4 implementation timeline (29 days) **Use Cases:** - Building MCP servers for Claude Desktop - Integrating mental models into AI agents - Edge database design - Server-side TypeScript architecture **Trigger Keywords:** MCP server, Model Context Protocol, Claude Desktop, MCP integration, npx server --- ### 2. SITREP Coordinator **File:** \`sitrep-coordinator/\` **Lines:** 645 **Purpose:** Military-style Situation Reports for multi-agent coordination **Capabilities:** - Structured SITREP generation - Authorization code management - Multi-agent handoff protocols - Status tracking () - Escalation frameworks - 4 coordination patterns **Use Cases:** - Multi-agent project coordination - Status reporting and updates - Task handoffs between agents - Blocker escalation - Team communication **Trigger Keywords:** SITREP, situation report, status update, multi-agent, coordination, handoff --- ### 3. GitHub Repository Architect **File:** \`github-repository-architect/\` **Lines:** 730 **Purpose:** Complete GitHub repository setup with production standards **Capabilities:** - Repository initialization (30min-3hr) - Community health files (README, LICENSE, CONTRIBUTING, etc.) - Issue and PR templates - CI/CD workflows (GitHub Actions) - GitHub Pages deployment - Branch protection and security **Use Cases:** - Creating new GitHub repositories - Setting up repository infrastructure - Automating quality checks - Documentation site deployment - Open-source project setup **Trigger Keywords:** GitHub repo, repository setup, CI/CD, GitHub Pages, community standards --- ### 4. HUMMBL Base120 Framework **File:** \`hummbl-framework/\` **Lines:** 239 **Version:** 1.0-beta (Definitive Reference) **Purpose:** Complete Base120 mental models reference with official codes, names, and one-line definitions **Capabilities:** - All 120 validated models with precise coding system - Perspective / Identity (P1-P20) - Inversion (IN1-IN20) - Composition (CO1-CO20) - Decomposition (DE1-DE20) - Recursion (RE1-RE20) - Meta-Systems (SY1-SY20) - Model selection guidance with quick reference table - One-line definitions for rapid agent coordination - Validation checklist to ensure correct usage - Complete provenance and source documentation **Use Cases:** - Mental model reference and application - Model selection for problem-solving - Transformation analysis - MCP server development (model knowledge) - Agent training and coordination - Strategic decision-making **Trigger Keywords:** HUMMBL, Base120, mental models, transformations, framework, P1-P20, IN1-IN20, CO1-CO20, DE1-DE20, RE1-RE20, SY1-SY20 --- ### 5. Transformation Workflow **File:** \`transformation-workflow/\` **Lines:** 720+ **Purpose:** Practical application guide for 6 HUMMBL transformations **Capabilities:** - 6 complete transformation workflows - When-to-use guides for each transformation - Output templates and formats - Combination patterns (P→DE→CO, etc.) - Real-world examples - Common pitfalls and solutions **Use Cases:** - Applying mental models to problems - Transformation selection guidance - Multi-model analysis - Structured problem-solving - Strategic thinking workflows **Trigger Keywords:** transformation, workflow, P/IN/CO/DE/RE/SY, mental model application --- ### 6. UI/UX Designer **File:** \`ui-ux-designer/\` **Lines:** 600+ **Purpose:** Modern UI/UX design for web and mobile applications **Capabilities:** - Design principles (visual hierarchy, consistency, affordance) - User research and testing - Interaction patterns - Accessibility (WCAG 2.1) - Responsive design (mobile-first) - Modern frameworks (Tailwind CSS, shadcn/ui, Material UI) - Complete workflow (research → design → implementation → testing) - Code examples and patterns **Use Cases:** - Web application UI design - Mobile app interfaces - Design system creation - Accessibility compliance - Responsive layouts - Component libraries **Trigger Keywords:** UI/UX, design, interface, accessibility, responsive, Tailwind, user experience --- ## Installation ### Claude Code 1. Clone this repository: \`\`\`bash git clone https://github.com/hummbl-dev/hummbl-claude-skills.git \`\`\` 2. Symlink to Claude Code skills directory: \`\`\`bash ln -s /path/to/hummbl-claude-skills ~/.claude/skills/hummbl \`\`\` 3. Verify skills loaded: \`\`\`bash claude skills list \`\`\` ### Claude API Skills can be loaded via the Agent SDK or directly in your application: \`\`\`typescript import \{ Skill \} from '@anthropic-ai/sdk/skills'; const mcpSkill = await Skill.load('path/to/mcp-server-developer/SKILL.md'); \`\`\` ### Claude.ai (if supported) Follow [Anthropic's documentation](https://docs.claude.com/en/docs/agents-and-tools/agent-skills) for using skills on Claude.ai. ## Usage Examples ### Example 1: Building an MCP Server \`\`\`typescript // With MCP Server Developer skill loaded const prompt = \` Create an MCP server that exposes mental model transformations. Include: TypeScript, D1 database, 3 endpoints (perspective, decompose, synthesize) \`; \`\`\` The skill will guide through: - Project scaffolding - Database schema - Endpoint implementation - Telemetry setup - Testing and deployment ### Example 2: Multi-Agent Coordination \`\`\`typescript // With SITREP Coordinator skill loaded const prompt = \` Generate a SITREP for our MCP server project. Status: Week 1 complete, 3 endpoints done, database operational. Auth code: HUMMBL-MCP-WEEK1-001 \`; \`\`\` Output: Military-style structured status report with completed/in-progress/blocked sections. ### Example 3: Problem Analysis \`\`\`typescript // With HUMMBL Framework + Transformation Workflow skills loaded const prompt = \` Analyze this product decision using HUMMBL transformations: Should we build feature X or feature Y? Apply Perspective, Decomposition, and Meta-Systems transformations. \`; \`\`\` Result: Multi-perspective analysis with stakeholder mapping, component breakdown, and systems thinking. ## Skill Quality All skills meet production quality standards: | Criterion | Target | Status | |-----------|--------|--------| | Format Compliance | 100% Anthropic spec | | | Comprehensiveness | >300 lines | | | Real Examples | 2+ per skill | | | Templates | Included | | | Best Practices | Documented | | | Common Pitfalls | Addressed | | **Overall Quality:** 10/10 (Context Engineering validation) ## Development ### Skill Structure Each skill follows this structure: \`\`\` skill-name/ ├── SKILL.md # Main skill file with YAML frontmatter └── README.md # Skill-specific documentation \`\`\` **SKILL.md Format:** \`\`\`markdown --- name: skill-name description: What this skill does and when to use it --- # Skill Name [Comprehensive guidance, templates, examples] \`\`\` ### Contributing We welcome contributions! To add a new skill: 1. Fork this repository 2. Create skill directory with SKILL.md and README.md 3. Follow Anthropic .skill format specification 4. Include real-world examples (minimum 2) 5. Document common pitfalls 6. Submit pull request See [CONTRIBUTING.md](CONTRIBUTING.md) for detailed guidelines. ## Related Projects - **[HUMMBL Unified Tier Framework](https://github.com/hummbl-dev/HUMMBL-Unified-Tier-Framework)** - Problem complexity classification and learning progression framework - **HUMMBL MCP Server** (Coming Soon) - MCP server implementation using these skills - **HUMMBL API** (Planned) - RESTful API for mental models ## Resources - **Documentation:** https://hummbl-dev.github.io/HUMMBL-Unified-Tier-Framework/ - **Anthropic Skills Guide:** https://docs.claude.com/en/docs/agents-and-tools/agent-skills - **MCP Specification:** https://spec.modelcontextprotocol.io/ - **HUMMBL Website:** https://hummbl.io ## License MIT License - see [LICENSE](LICENSE) file for details. ## Authors **HUMMBL, LLC** - Framework Design: HUMMBL Research Team - Skills Development: Multi-Agent Development Team - Validation: Context Engineering Agent - Quality Assurance: 10/10 production standard ## Citation If you use these skills in your research or project, please cite: \`\`\`bibtex @software\{hummbl_claude_skills, title = \{HUMMBL Claude Skills\}, author = \{HUMMBL, LLC\}, year = \{2025\}, url = \{https://github.com/hummbl-dev/hummbl-claude-skills\}, version = \{1.0.0\} \} \`\`\` ## Changelog See [CHANGELOG.md](CHANGELOG.md) for version history. ## Support For issues, questions, or feedback: - **Issues:** https://github.com/hummbl-dev/hummbl-claude-skills/issues - **Discussions:** https://github.com/hummbl-dev/hummbl-claude-skills/discussions --- **Built with ️ by HUMMBL for the Claude AI community**