{"id":48596149,"url":"https://github.com/emminix/echocontext-factory","last_synced_at":"2026-04-08T21:01:59.653Z","repository":{"id":304978139,"uuid":"1021491684","full_name":"EmminiX/EchoContext-Factory","owner":"EmminiX","description":"🎵 Voice-enabled context engineering system for Claude Code with accessibility features and documentation-to-development bridge","archived":false,"fork":false,"pushed_at":"2025-10-29T23:50:20.000Z","size":1365,"stargazers_count":5,"open_issues_count":1,"forks_count":1,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-30T01:28:44.921Z","etag":null,"topics":["accessibility","automation","claude-code","context-engineering","developer-tools","elevenlabs","hooks","javascript","mcp-integration","multi-agent","openai","python","tts","typescript","voice-ai","workflow"],"latest_commit_sha":null,"homepage":"https://emmi.zone","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/EmminiX.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-07-17T13:29:30.000Z","updated_at":"2025-10-04T07:02:50.000Z","dependencies_parsed_at":"2025-07-17T19:13:32.706Z","dependency_job_id":"736a06d8-bcea-41ff-9cc5-17ff479873c0","html_url":"https://github.com/EmminiX/EchoContext-Factory","commit_stats":null,"previous_names":["emminix/echocontext-factory"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/EmminiX/EchoContext-Factory","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EmminiX%2FEchoContext-Factory","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EmminiX%2FEchoContext-Factory/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EmminiX%2FEchoContext-Factory/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EmminiX%2FEchoContext-Factory/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/EmminiX","download_url":"https://codeload.github.com/EmminiX/EchoContext-Factory/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/EmminiX%2FEchoContext-Factory/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31573788,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-08T14:31:17.711Z","status":"ssl_error","status_checked_at":"2026-04-08T14:31:17.202Z","response_time":54,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["accessibility","automation","claude-code","context-engineering","developer-tools","elevenlabs","hooks","javascript","mcp-integration","multi-agent","openai","python","tts","typescript","voice-ai","workflow"],"created_at":"2026-04-08T21:01:13.179Z","updated_at":"2026-04-08T21:01:59.643Z","avatar_url":"https://github.com/EmminiX.png","language":"JavaScript","readme":"# 🎵 EchoContext Factory\n\n\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"assets/EMMI.png\" alt=\"EMMI - Engaging Minds, Merging Ideas\" width=\"300\"\u003e\n\u003c/div\u003e\n\n**Turn Claude Code into your personal AI assistant with voice guidance and zero prompt engineering**\n\n---\n\n## 🎯 Why Use This?\n\nTransform your development workflow with:\n\n- **🎵 Voice guidance** - Your AI talks to you throughout the process\n- **🧠 Zero prompt engineering** - Just answer simple questions, no complex prompts needed  \n- **📝 Instant documentation** - Auto-generates comprehensive project docs with live web research\n- **♿ Neurodivergent-friendly** - One question at a time, clear structure, predictable flow\n- **🚀 Multi-agent coordination** - Complex tasks handled by specialized AI agents in parallel\n\n**Perfect for:** Developers who want engaging, accessible AI assistance without the complexity.\n\n---\n\n## 🚀 Quick Installation (3 Steps)\n\n### Step 1: Download and Install\n\n**All Platforms (Linux, macOS, Windows):**\n```bash\n# Clone or download this repository\ngit clone https://github.com/EmminiX/EchoContext-Factory.git\ncd EchoContext-Factory\n\n# Run the cross-platform installer\nchmod +x install.sh\n./install.sh\n```\n\nThe installer automatically detects your platform and configures everything appropriately!\n\n### Step 2: Add Your API Keys (Optional but Recommended)\n```bash\n# Edit the environment file (path shown by installer)\nnano ~/.claude/.env\n\n# Add your keys for enhanced voice features:\nELEVENLABS_API_KEY=your_key_here     # Best voice quality\nOPENAI_API_KEY=your_key_here         # Good voice quality  \nENGINEER_NAME=YourName               # Personalizes voice messages\n```\n\n**Platform-specific notes:**\n- **Linux**: Excellent support, [install espeak](https://espeak.sourceforge.net/download.html) for better fallback TTS\n- **macOS**: Full support with built-in TTS. Enable \"Announce notifications\" in System Settings → Accessibility → Spoken Content for best experience\n- **Windows**: Core features work, may need [WSL](https://docs.microsoft.com/en-us/windows/wsl/install) for advanced Python hooks\n\n### Step 3: Start Building!\n```bash\n# In Claude Code, type any of these commands:\n/start-project      # Interactive project setup with voice guidance\n/voice-status       # Test that voice is working\n```\n\n**That's it!** Your AI assistant is now voice-enabled and ready to guide you.\n\n### 🔧 Recommended: Enhanced Research Capabilities\n\nFor the best experience with multi-agent coordination and research, install these MCP providers:\n\n- **Import from Claude Desktop**: `claude mcp add-from-claude-desktop`  \n- **Essential MCPs**: [Context7 MCP](https://modelcontextprotocol.io/quickstart/user), [Perplexity MCP](https://github.com/jsonallen/perplexity-mcp), [Tavily MCP](https://github.com/tavily-ai/tavily-mcp)\n\n*Note: Context7 and Perplexity MCPs are MANDATORY for all specialized agents, enabling comprehensive codebase analysis, current best practices research, and cross-validated findings for accuracy*\n\n### 🤖 Claude Code Agents Support\n\nEchoContext Factory leverages Claude Code's sub-agent system for 10-20x speed improvements:\n\n- **Pre-configured Agents**: 10 specialized agents in `.claude/agents/` for research, implementation, testing, and documentation\n- **Parallel Execution**: Multiple agents work simultaneously on different aspects of your project\n- **Smart Orchestration**: Agents coordinate based on task dependencies with sophisticated task decomposition\n- **Automatic Fallback**: Works normally if agents aren't available (non-Claude Code environments)\n\n---\n\n## 🎵 What Makes This Special?\n\n### 🗣️ Your AI Assistant Talks to You\n\nNever feel alone while coding! Your assistant provides:\n\n- **🎯 Progress updates** - \"Creating your documentation...\"\n- **✅ Confirmations** - \"Perfect! Moving to the next step\"  \n- **🎉 Celebrations** - \"Amazing work! Your project is ready!\"\n- **❓ Gentle guidance** - \"What type of project are you building?\"\n\n**Works 3 ways**: Premium voice (ElevenLabs) → Good voice (OpenAI) → Computer voice (always works)\n\n### 🧠 Zero Mental Overhead\n\n**Before**: Spend hours crafting the perfect prompt  \n**After**: Just answer simple questions like \"What's your project about?\"\n\n- **Adaptive flow**: Starts with your project description, then asks smart follow-ups\n- **Maximum 9 questions**: Never overwhelming, usually fewer based on your needs  \n- **Context-aware**: Each question builds on your previous answers\n- No prompt engineering needed\n\n### 📋 What You Get\n\nEvery project automatically generates:\n\n```\n📁 Your Project\n├── 📄 CLAUDE.md         # Complete AI assistant context\n├── 📋 PRD.md           # Professional requirements doc  \n├── ✅ TASKS.md         # Step-by-step implementation guide\n└── 🔗 Live web research # Current best practices included\n```\n\n**Real example**: Answer up to 9 adaptive questions → Get 3 comprehensive documents with current industry standards\n\n---\n\n## 🎵 Voice Hooks System\n\n### 🔊 How Voice Works\n\nEchoContext Factory includes a sophisticated **3-tier voice fallback system** that ensures you always get voice announcements, regardless of your API key setup:\n\n#### **Tier 1: Premium Voice (ElevenLabs)**\n- **Requires**: `ELEVENLABS_API_KEY` in your `.env` file\n- **Quality**: Professional-grade, natural-sounding voice (Charlotte voice, Eleven Turbo v2.5)\n- **Best for**: Daily development work where voice quality matters\n\n#### **Tier 2: Good Voice (OpenAI)**  \n- **Requires**: `OPENAI_API_KEY` in your `.env` file\n- **Quality**: Clear, pleasant AI-generated voice (Nova voice, TTS-1 model)\n- **Best for**: Regular use when ElevenLabs isn't available\n\n#### **Tier 3: System Voice (Always Works)**\n- **Requires**: Nothing! Works out of the box on all platforms\n- **Quality**: Your computer's built-in text-to-speech\n- **Platform Support**: Windows SAPI, macOS NSSpeechSynthesizer, Linux espeak/festival\n- **Best for**: Immediate functionality without any setup or API keys\n\n### 🔧 Voice Hooks Architecture\n\nThe voice system uses **Python hooks** that automatically trigger during key development moments:\n\n#### **Hook Types \u0026 Triggers**\n- **Factory Notifications** (`factory_notification.py`): Triggered by TodoWrite events during /start-project phases\n- **General Notifications** (`notification.py`): Announced when Claude needs your input or guidance\n- **Session Completion** (`stop.py`): Celebrates successful completion of work sessions\n- **Agent Coordination** (`subagent_stop.py`): Reports when multi-agent tasks complete\n\n#### **What You Actually Hear**\n```python\n# Real voice announcements during development:\n📝 Phase transitions     # \"EchoContext Factory is spinning up for maximum speed!\"\n🎯 Project completion    # \"Your comprehensive project documentation is ready!\"\n🤖 Agent completion     # \"Multi-agent coordination complete - results aggregated!\"\n⚠️ Input needed         # \"Your cognitive enhancement system needs direction\"\n✅ Final celebration    # \"Mission accomplished! Ready for your next challenge!\"\n```\n\n**Note**: Voice announcements happen **after** key actions complete, not during interactive question flows.\n\n### 🎯 Personalized Announcements\n\nWhen you set `ENGINEER_NAME=YourName` in your `.env` file:\n- **70% of messages** become personalized: *\"Hey Sarah, time to sync minds\"*\n- **30% stay generic** for variety: *\"Your AI collaborator seeks your wisdom\"*\n- **Smart context**: Messages adapt to your workflow and progress\n\n### 🎚️ Voice Control\n\n```bash\n/voice-status      # Test voice system and check configuration\n/voice-toggle      # Turn voice announcements on/off instantly\n```\n\n**No API Keys? No Problem!** The system voice (Tier 3) ensures you get announcements even without any API keys configured.\n\n### 🎚️ Voice Customization\n\n#### **🏷️ Personal Name Integration**\n\nAdd your name for personalized announcements:\n```bash\n# In ~/.claude/.env\nENGINEER_NAME=Sarah\n```\n\n**Name Usage Frequency:**\n- **70% of messages** include your name: *\"Hey Sarah, time to sync minds\"*\n- **30% stay generic** for variety: *\"Your AI collaborator seeks your wisdom\"*\n- **Smart context**: Name usage adapts to workflow phases and progress\n\n#### **🔊 Voice Quality Tiers**\n\n**ElevenLabs (Premium):**\n```bash\n# In ~/.claude/.env  \nELEVENLABS_API_KEY=your_key_here\n```\n- Professional voice actors, natural speech patterns\n- Multiple voice options, emotion and pace control\n\n**OpenAI (Good):**\n```bash\n# In ~/.claude/.env\nOPENAI_API_KEY=your_key_here\n```\n- Clear AI-generated voice, reliable quality\n- Multiple voice models available\n\n**System Voice (Always Works):**\n- No configuration needed\n- Uses your computer's built-in TTS (macOS `say`, Windows Speech API, Linux `espeak`)\n- Immediate functionality without any setup\n\n#### **🎛️ Voice Controls**\n\n```bash\n/voice-status      # Test current voice setup and check all tiers\n/voice-toggle      # Instantly turn voice announcements on/off\n```\n\n**Voice automatically cascades**: ElevenLabs → OpenAI → System Voice if keys aren't available.\n\n---\n\n## 🚀 Main Commands\n\n### 🎯 Start a New Project\n```bash\n/start-project\n```\n\n**What happens**: Conducts an intelligent interview using an adaptive 9-question system, then generates comprehensive project documentation with live MCP research.\n\n#### **🧠 Adaptive Question System**\n\n**Smart Flow Design:**\n- **Question 1-3**: Core project foundation (name, detailed description, type)\n- **Question 2 is KEY**: Your detailed project description drives all subsequent question selection\n- **Questions 4-9**: Intelligently selected based on keywords in your description\n- **Maximum 9 questions**: Reduced from 18 for better neurodivergent accessibility\n- **Progress tracking**: Clear 11%, 22%, 33%... progression to 100%\n\n**Keyword-Based Adaptation:**\n```javascript\n// Example: If your description contains \"mobile app\", \"social features\", \"user authentication\"\n// The system automatically selects relevant questions about:\n- Mobile platform preferences (iOS/Android)\n- Social interaction features \n- Authentication \u0026 security requirements\n- Performance \u0026 scalability needs\n```\n\n#### **🔍 Comprehensive MCP Research (v2.5.0)**\n\n**Mandatory Research Integration:**\n- **15+ intelligent queries** generated per project based on your tech stack\n- **Context7 MCP**: Comprehensive codebase analysis and documentation research\n- **Perplexity MCP**: Current industry standards and expert best practices  \n- **Tavily MCP**: Real-time web research for latest trends and techniques\n- **Cross-validation**: All MCP sources verify findings for accuracy confidence scoring\n\n#### **🤖 Multi-Agent Research Teams**\n\nWhen Claude Code agents are available, `/start-project` spawns **4 specialized research agents** working in parallel:\n\n**Research Agent Team:**\n- **Tech Stack Agent**: Deep-dives into best practices for your chosen technologies\n- **Architecture Agent**: Analyzes optimal design patterns for your project type\n- **Security Agent**: Investigates OWASP 2024/2025 compliance requirements\n- **Deployment Agent**: Explores modern infrastructure and DevOps strategies\n\n**Benefits:**\n- **10-20x faster research** through parallel execution\n- **Cross-validated findings** between agents and MCP tools\n- **Comprehensive coverage** of all project aspects simultaneously\n- **Research source attribution** in all generated documentation\n\n#### **📋 Generated Documentation**\n\n**Complete Project Context:**\n```\n📁 .claude/\n├── 📄 CLAUDE.md         # Complete AI assistant context with research links\n├── 📋 PRD.md           # Professional requirements doc with industry standards\n├── ✅ TASKS.md         # Step-by-step implementation guide with best practices\n└── 🔗 Research sources # All MCP research properly attributed\n```\n\n#### **📝 Text-Based Interactive Flow**\n\n**Example Project Setup:**\n1. **Question 1/9**: \"What's your project called?\" *(11% complete)*\n2. **You**: \"My Recipe App\" \n3. **Question 2/9**: \"Please provide a detailed description...\" *(22% complete)*\n4. **You**: \"A web app where users can save, organize, and share their favorite recipes with social features...\"\n5. 🧠 *AI analyzes keywords: \"web app\", \"social features\", \"organize\"*\n6. **Question 3/9**: \"What type of users will primarily use this app?\" *(33% complete)*\n7. **Questions 4-9**: *Adaptively selected based on your project description*\n8. 🎯 **Progress**: \"Great! 100% complete. Launching research agents...\"\n9. 🤖 *4 research agents work in parallel using MCP tools*\n10. 🔍 *15+ research queries execute across Context7, Perplexity, and Tavily*\n11. ✅ *Voice announcement: \"Your comprehensive project documentation is ready!\"*\n\n### 🧠 Generate or Improve Prompts\n```bash\n/promptsage\n```\n**What happens**: Uses the PromptSage framework to create new system prompts or improve existing ones with model-specific XML layering.\n\n**Two modes**: Create new expert assistant prompts or enhance your existing prompts. Automatically applies optimal XML structure depth based on the target AI model (Claude: 3 layers, GPT/Gemini: 2 layers, Mistral/Llama: 1 layer).\n\n**Example**: \"Create a Python tutor prompt for Claude\" → Get a perfectly structured system prompt optimized for your chosen AI model.\n\n### 🚀 Development with Agents\n```bash\n/dev-agents\n```\n**What happens**: Spawns 5 specialized development agents to build your project in parallel using Claude Code's agent system.\n\n**Agent Team**: Frontend, Backend, Database, Testing, and Documentation agents work simultaneously on different aspects of your project using the pre-configured agent templates in `.claude/agents/`.\n\n**Benefits**: 10-20x faster development through parallel execution with smart task coordination and dependency management.\n\n**Requirements**: Only works in Claude Code (claude.ai/code) environment where the Task tool is available.\n\n### 🤖 Complex Task Help\n```bash\n/multiagent\n```\n**What happens**: Analyzes your task complexity and spawns specialized agents that work in parallel to research, analyze, and solve complex development challenges using mandatory MCP tools.\n\n**Real Parallel Execution**: Uses Claude Code's Task tool for true simultaneous agent execution with sophisticated task decomposition and dependency management.\n\n**Agent Types**: Automatically selects from Research, Architecture, Implementation, Testing, and Documentation specialists based on task requirements - each with access to Context7 and Perplexity MCP for comprehensive research.\n\n**Example**: \"Help me implement user authentication\" → System analyzes complexity, spawns Security and Implementation agents that research OWASP standards via MCP tools, generate secure code, and create testing strategies simultaneously.\n\n### 📋 Feature Requirements\n```bash\n/generate-prp\n```\n**What happens**: Creates detailed Product Requirements Prompts (PRPs) for specific features with current industry best practices and security considerations.\n\n**Process**: Interactive feature questions → Automated codebase analysis → Live MCP research → Professional requirements document saved to `generated-prps/` directory with full source attribution.\n\n**Example**: \"I need a login system\" → Get comprehensive requirements document with OWASP 2024 security standards, implementation details, testing strategies, and research-backed recommendations from Context7 and Perplexity sources.\n\n### 🔧 Utility Commands\n```bash\n/voice-status      # Test if voice is working and check configuration\n/voice-toggle      # Turn voice announcements on/off\n/agent-status      # Check Claude Code agent system status and availability\n```\n\n---\n\n## 🧠 Understanding CLAUDE.md\n\nThe `CLAUDE.md` file in your project contains a **comprehensive system prompt** that transforms Claude into a world-class development assistant with:\n\n### 🎯 What's Inside CLAUDE.md\n- **Core Identity**: Full-stack developer AND cybersecurity expert specializing in secure, accessible software\n- **Behavioral Directives**: Mandatory MCP tool usage (Context7, Perplexity) for all research tasks\n- **Security Framework**: OWASP Top 10 2024 integration and defense-in-depth principles\n- **Tech Stack 2025**: Modern frameworks, security tools, and accessibility standards\n- **Multi-Agent Coordination**: Smart work principles and specialized agent spawning requirements\n- **Accessibility-First**: Neurodivergent-friendly design patterns and WCAG 2.2 AA compliance\n\n### 💡 Pro Tip: Getting Unstuck\n\nWhen Claude seems stuck on complex issues, **add these instructions** along with your task:\n\n```\n- Take a step back and analyze the situation.\n- Don't complicate things.\n- Spawn specialized agents to help you gather information about the issue or do research.\n- Never assume anything, always document yourself using the MCP tools you have at your disposal.\n- Think deep and hard, be diligent, don't work hard work smart.\n- Look for patterns.\n- Create a good action plan.\n```\n\n**Why this works**: These instructions activate Claude's analytical mode and leverage the multi-agent system for complex problem-solving, exactly as defined in CLAUDE.md.\n\n### 🤖 Pro Tip: Automatic Agent Spawning\n\nFor complex tasks, simply tell Claude to **\"spawn multiple agents\"** and it will:\n\n- **Automatically decide** what specialist agents are needed\n- **Spawn the appropriate experts** (research, architecture, security, implementation, etc.)\n- **Assign detailed tasks** to each agent based on the problem\n- **Coordinate the work** and aggregate results\n\n**Example**: \"Help me build a secure user authentication system - spawn multiple agents\" → Claude spawns Security, Backend, Database, and Testing agents automatically.\n\n### 🔄 How It All Connects\n\n1. **CLAUDE.md** = Your AI assistant's \"personality\" and capabilities\n2. **EchoContext Factory** = The workflow system that uses CLAUDE.md\n3. **MCP Tools** = Research capabilities (Context7, Perplexity, Tavily)\n4. **Specialized Agents** = Expert assistants spawned based on CLAUDE.md specifications\n\n**Result**: A sophisticated AI development environment that thinks systematically, works securely, and builds accessibly.\n\n---\n\n## 🤔 Need Help?\n\n### 🤖 Agents Not Working?\n\n**IMPORTANT**: Agents ONLY work inside Claude Code (claude.ai/code), not in terminal/VS Code.\n\n**Quick diagnosis:**\n1. Run `/agent-status` to check availability\n2. Look for \"Claude Code agent system detected\" message\n3. If you see \"Agent system not available\", you're in fallback mode\n\n**Solution**: Open this project in Claude Code at claude.ai/code to use agents.\n\n### 🔊 Voice Not Working?\n\n**Try these in order:**\n\n1. **Test it**: `/voice-status` \n2. **No sound?** Check your computer volume\n3. **Want better voice?** Add API keys to `~/.claude/.env`:\n   ```bash\n   ELEVENLABS_API_KEY=your_key_here     # Best quality\n   OPENAI_API_KEY=your_key_here         # Good quality\n   ```\n4. **Still nothing?** Computer voice works without any setup\n\n### 📁 Installation Issues?\n\n**Problem**: \"Permission denied\" or \"command not found\"  \n**Solution**: Make sure you're in the EchoContext-Factory folder when running `bash install.sh`\n\n**Problem**: \"No such file or directory\"  \n**Solution**: Your `.claude` folder might be elsewhere. Try: `find ~ -name \".claude\" -type d`\n\n**Problem**: Commands don't work in Claude Code  \n**Solution**: Restart Claude Code after installation\n\n**Problem**: Installation script fails or doesn't work properly  \n**Solution**: Manual installation (always works):\n```bash\n# 1. Backup your existing .claude folder first!\ncp -r ~/.claude ~/.claude.backup.$(date +%Y%m%d_%H%M%S)\n\n# 2. Copy all files manually from the downloaded repo to ~/.claude/\ncp -r EchoContext-Factory/commands ~/.claude/\ncp -r EchoContext-Factory/config ~/.claude/\ncp -r EchoContext-Factory/data ~/.claude/\ncp -r EchoContext-Factory/hooks ~/.claude/\ncp -r EchoContext-Factory/lib ~/.claude/\ncp -r EchoContext-Factory/templates ~/.claude/\ncp EchoContext-Factory/CLAUDE.md ~/.claude/\ncp EchoContext-Factory/settings.json ~/.claude/\ncp EchoContext-Factory/settings.fallback.json ~/.claude/\ncp EchoContext-Factory/.env.sample ~/.claude/\n\n# 3. Create .env file if needed\ncp ~/.claude/.env.sample ~/.claude/.env\n```\n\n**Problem**: Error \"uv: not found\" when hooks run  \n**Solution**: Either install uv or use fallback settings:\n```bash\n# Option 1: Install uv (recommended)\ncurl -LsSf https://astral.sh/uv/install.sh | sh\n\n# Option 2: Use system python fallback\ncd ~/.claude\ncp settings.fallback.json settings.json\n```\n\n---\n\n## 🎉 That's It!\n\nYou now have a voice-enabled AI assistant that will guide you through every step of development. \n\n### 🚀 Ready to Start?\n\n1. **Test voice**: `/voice-status` \n2. **Create your first project**: `/start-project`\n3. **Need complex help**: `/multiagent`\n\n### 🎵 Remember\n\n- Your AI talks to you (no more feeling alone while coding!)\n- One simple question at a time (never overwhelming)\n- Everything works without API keys (computer voice is always available)\n- All commands are designed to be neurodivergent-friendly\n\n**Happy building!** 🏭✨\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions! \n\n**How to contribute:**\n1. Fork the repository\n2. Create a feature branch (`git checkout -b feature/amazing-feature`)\n3. Make your changes following the accessibility-first principles\n4. Test with voice features and ensure neurodivergent-friendly design\n5. Submit a pull request\n\n---\n\n**Version**: 2.5.0 | **Created by**: [Emmi C.](https://emmi.zone) | **License**: MIT\n\n### 🆕 What's New in v2.5.0?\n\n- **🧠 Smarter Question Flow**: New adaptive 9-question flow that starts with your project description and intelligently selects follow-up questions\n- **🔍 Enhanced MCP Integration**: All multi-agent tasks now mandate Context7 and Perplexity MCP usage for comprehensive research and validation\n- **⚡ Reduced Cognitive Load**: Fewer, smarter questions that adapt to your project type (maximum 9, often fewer)\n- **🎯 Better Context Building**: First question captures your project vision, remaining questions build context intelligently\n- **🤖 Claude Code Agents**: Full integration with Claude Code's sub-agent system for parallel execution and 10-20x speed improvements\n- **📁 Pre-configured Agents**: 10 specialized agent definitions ready to use in `.claude/agents/`\n- **🚀 Agent Teams**: Development teams for /dev-agents and research teams for /start-project\n\n*Accessibility improvement: Less overwhelming question flow perfect for neurodivergent developers*\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Femminix%2Fechocontext-factory","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Femminix%2Fechocontext-factory","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Femminix%2Fechocontext-factory/lists"}