{
  "markdown": "# TCAM Memory System\n\nResilient 4-Layer Memory Architecture for TCAM v1.4 (Triadic Cognitive Augmentation Model)\n\n## Overview\n\nThe TCAM Memory System implements a fault-tolerant memory architecture where each layer operates independently, ensuring that system failures in one layer do not cascade to others. Memory operations are handled by a dedicated Memory Service running on Qwen 3.5 9B (local LLM).\n\n## Architecture\n\n### Four Independent Layers\n\n- **L1: Chronicle** - Immutable historical record (file system, markdown)\n- **L2: Active Stream** - Volatile working memory (Cloud LLM context + Redis)\n- **L3: Hive Mind** - Semantic memory (Qdrant vector DB + Mem0)\n- **L4: Agent Codex** - Personal knowledge base (file system, markdown)\n\n### Memory Service\n\nDedicated process running on Qwen 3.5 9B for:\n- Truth extraction from dialogue\n- Chronicle inscription\n- Hive Mind indexing\n- Agent Codex updates\n- Sleeping cycle orchestration\n\n## Technology Stack\n\n- **TypeScript/Node.js** - Core implementation\n- **Qwen 3.5 9B** - Local LLM for memory operations\n- **Mem0** - Automatic fact extraction\n- **Qdrant** - Vector database for semantic search\n- **Redis** - Fast state persistence (LangGraph checkpointer)\n- **LangGraph** - Multi-agent orchestration\n- **Jest** - Testing framework\n- **fast-check** - Property-based testing\n\n## Project Structure\n\n```\n.\n├── src/                    # Source code\n├── tests/                  # Test files\n├── data/                   # Chronicle storage\n├── codex/                  # Agent Codex (L4)\n├── docs/                   # Documentation\n└── .kiro/specs/           # Specification documents\n```\n\n## Getting Started\n\n### Prerequisites\n\n- Node.js 18+\n- Ollama (for Qwen 3.5 9B)\n- Docker (for Qdrant and Redis)\n\n### Installation\n\n```bash\n# Install dependencies\nnpm install --legacy-peer-deps\n\n# Pull Qwen 3.5 9B model\nollama pull qwen2.5:9b-instruct-q4_K_M\n\n# Pull embedding model\nollama pull nomic-embed-text\n\n# Start Qdrant (using setup script)\n# Windows PowerShell:\n.\\scripts\\setup-qdrant.ps1\n\n# Linux/Mac:\nbash scripts/setup-qdrant.sh\n\n# Or manually with Docker:\ndocker run -d --name qdrant -p 6333:6333 -p 6334:6334 -v qdrant_storage:/qdrant/storage qdrant/qdrant\n\n# Start Redis\ndocker run -d --name redis -p 6379:6379 redis\n\n# Build and install globally\nnpm run build\nnpm install -g . --legacy-peer-deps\n```\n\n### CLI Usage\n\n```bash\n# Interactive main menu\nanots\n\n# Real-time monitoring dashboard (Terminal UI)\nanots dashboard\n\n# Chat with Axiom (TCAM Node C)\nanots axiom\n\n# Start MCP server (19 tools for IDE integration)\nanots mcp:start\n\n# Start REST API server\nanots api:start\n\n# Start API with Axiom chat endpoint\nanots api:start --axiom\n\n# Search memory\nanots memory:search \"your query\"\n\n# Import conversation (JSON/Markdown)\nanots import conversation.json --type general\n\n# System status\nanots status\n\n# Configuration wizard\nanots setup\n\n# Show all commands\nanots --help\n```\n\n### Dashboard Features\n\nThe Terminal UI dashboard (`anots dashboard`) provides:\n- Real-time memory layer health monitoring\n- System statistics (chapters, sessions, memories)\n- Activity log with timestamps\n- Keyboard shortcuts (F1-F3, R, Q)\n- Auto-refresh every 2 seconds\n- Cyberpunk aesthetic with 90s BBS vibes\n\n### Verify Setup\n\n```bash\n# Check Qdrant\ncurl http://localhost:6333/health\n\n# Check Redis\nredis-cli ping\n\n# Check Ollama\nollama list\n```\n\n### Development\n\n```bash\n# Build\nnpm run build\n\n# Run tests\nnpm test\n\n# Run tests with coverage\nnpm run test:coverage\n\n# Lint\nnpm run lint\n\n# Format code\nnpm run format\n```\n\n## Importing Conversations\n\nImport large conversation files into the ANOTS memory system:\n\n```bash\n# Import Ubik conversation\nnpm run import -- data/import/conversation.json --type ubik\n\n# Import Axiom conversation\nnpm run import -- data/import/conversation.md --type axiom\n\n# Preview import without writing (dry run)\nnpm run import -- data/import/conversation.json --dry-run\n\n# Custom chunk size (messages per chapter)\nnpm run import -- data/import/large-file.json --type ubik --chunk-size 100\n```\n\n### What Gets Imported\n\n1. **Chronicle**: Conversation history in chapters\n2. **Hive Mind**: Semantic indexing for search\n3. **Codex**: Extracted truths and insights\n\nSee `data/import/README.md` for detailed documentation and examples.\n\n## Specification\n\nSee `.kiro/specs/memory-system/` for complete specification:\n- `requirements.md` - Functional requirements\n- `design.md` - Technical design\n- `tasks.md` - Implementation tasks\n\n## License\n\nMIT\n",
  "bytes": 4497,
  "sha": "6167240ec98481d1f519e0d141c0597e1aca702561a13b6dcbd74e450298aadf",
  "repo_slug": "patheras/the-anots",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_anots_directory_410873db/readme"
}