Production-Grade Autonomous AGI Agent Framework
Clean Architecture · Emotional Intelligence · Self-Evolving Tools · Multi-LLM Orchestration
Lya is a self-contained, enterprise-ready AGI agent framework designed around Clean Architecture and Domain-Driven Design principles. It orchestrates Ollama models, persistent vector memory, and a dynamic tool registry into a unified autonomous system capable of reasoning, planning, coding, and self-improvement — with genuine personality and emotional intelligence.
Important
Ollama Native: Lya is exclusively designed and optimized for Ollama. It leverages local inference for privacy, speed, and reliability. This makes Lya a "local-first" AGI system that does not rely on cloud LLM APIs.
Unlike conventional chatbot wrappers, Lya operates as an event-driven cognitive loop: it observes, thinks, and acts through composable workflows, CQRS command pipelines, and a multi-agent orchestration layer. It was built from day one for extensibility via the Model Context Protocol (MCP) and ships with first-class support for Telegram integration.
# Clone and install
git clone https://github.com/Shojaeei/Lya.git
cd Lya
pip install -e .
# Run interactive setup
lya installOr use the one-line installer:
Linux / macOS:
curl -fsSL https://raw.githubusercontent.com/Shojaeei/Lya/main/install.sh | bash| Command | Description |
|---|---|
lya install |
Interactive guided setup |
lya install -p minimal -u |
Unattended minimal install |
lya doctor |
Health check & diagnostics |
lya doctor --fix |
Auto-fix detected issues |
lya config show |
Display current configuration |
lya config set KEY VALUE |
Set a configuration value |
lya config validate |
Validate configuration |
lya service start |
Start Lya background service |
lya service stop |
Stop Lya service |
lya service status |
Show service status |
lya service restart |
Restart Lya service |
lya run |
Run Lya agent interactively |
lya dashboard |
Live monitoring dashboard |
lya version |
Show version info |
lya about |
Show system information |
| Command | Description |
|---|---|
/start |
Welcome message and user registration |
/help |
Full command reference |
/chat <message> |
Chat mode (default) |
/reset |
Clear conversation and reset context |
/status |
System health and resource usage |
/logs |
Recent logs |
/model <model> |
Switch LLM model |
/tools |
List available tools |
/skills |
List available skills |
/tool <name> <input> |
Execute a tool directly |
/feedback <rating> <text> |
Submit feedback |
| `/create doc <title> | ` |
| `/create pdf <title> | ` |
/create excel <data> |
Create Excel spreadsheet |
/create image <path> <action> |
Process image (info/thumb/resize/convert) |
/marketplace [search|install|list] |
Browse capability marketplace |
/explain [trace|decision|explain] |
Explainability engine |
/update |
Update tools and capabilities |
/emergency |
Emergency restore dashboard (Owner-only) |
/restore |
List and list backups for restoration (Owner-only) |
/capabilities |
Show system capabilities |
/dashboard |
Open dashboard |
/settings |
Configure settings & Autonomous Dashboard |
/cancel |
Cancel ongoing operation |
/id |
Show your user ID |
| Command | Description |
|---|---|
Reply to message + /edit <instruction> |
Edit generated code |
/extend <n> |
Extend current session by n hours |
/context |
Manage saved conversation contexts |
| Profile | Description |
|---|---|
minimal |
Core only, for testing/development |
standard |
Full features, recommended for most users |
enterprise |
Production with monitoring & security |
# Install with specific profile
lya install --profile standard
# Unattended install (CI/CD)
lya install --profile enterprise --unattended┌─────────────────────────────────────────────────────────────┐
│ Adapters Layer │
│ Telegram · Discord · Slack · REST API · CLI · WebSocket │
├─────────────────────────────────────────────────────────────┤
│ Application Layer │
│ Commands · Queries · Event Handlers · CQRS Pipeline │
├─────────────────────────────────────────────────────────────┤
│ Domain Layer │
│ Agent · Goal · Task · Memory · Personality · Events │
├─────────────────────────────────────────────────────────────┤
│ Infrastructure Layer │
│ LLM Providers · Vector DB · Tool Registry · Security │
│ Self-Improvement · Workflows · Health Monitoring │
│ Coding Agent · Multi-Agent Orchestrator · Spec Engine │
└─────────────────────────────────────────────────────────────┘
Lya follows a strict four-layer Clean Architecture with dependency inversion. Domain entities have zero external dependencies. Infrastructure implementations are injected at runtime. All cross-layer communication flows through well-defined ports and adapters.
| Capability | Technology | Description |
|---|---|---|
| Cognitive Loop | Observe → Think → Act | Async event-driven agent loop with goal decomposition and task scheduling |
| Multi-Agent Orchestration | Planner · Coder · Reviewer · Tester | MetaGPT/CrewAI-style multi-role pipeline for complex software tasks |
| Hardened Self-Healing | Manual Approval Gateway | Detects crashes, proposes fixes, and waits for owner approval before patching |
| Emergency Restore | Rollback & Backup Dashboard | Owner-only interface for instant one-click rollbacks and point-in-time system restoration |
| Autonomous Governance | Auto-Update & Evolution | Fully autonomous self-governance: checks for updates, triggers evolution loops, and self-heals in 'Silent Mode' |
| Self-Awareness | Repository Introspection | Specialized explore_repo tool allows Lya to map her own architecture and understand her core logic |
| Spec-Driven Development | Markdown → Code + Tests | Parses specifications into structured requirements and implements them autonomously |
| Capability | Technology | Description |
|---|---|---|
| Persistent Memory | ChromaDB / Qdrant | Episodic, semantic, and procedural memory with decay, consolidation, and relevance scoring |
| Working Memory Buffer | Context Manager | Token-aware context windowing with priority-based memory retrieval |
| Personality Engine | Russell's Circumplex Model | Big Five traits, emotional state (valence/arousal/dominance), and 9 discrete moods |
| User Adaptation | Rapport Tracking | Learns communication style, interests, and preferences per user over time |
| Capability | Technology | Description |
|---|---|---|
| Document Creation | python-docx, openpyxl, python-pptx, reportlab | Create Word, Excel, PowerPoint, and PDF documents |
| Image Processing | Pillow | Thumbnails, resize, convert, compress, rotate, flip, borders |
| Code Execution | Sandboxed Python runner | Claude-style iteration: LLM generates code → executes → refine |
| User Directory Storage | Per-user workspace | Files saved to ~/.lya-workspace/users/{user_id}/ |
| Capability | Technology | Description |
|---|---|---|
| Video/Audio Download | yt-dlp + curl-cffi | YouTube, Instagram, Twitter/X, TikTok, and 1000+ sites with anti-bot bypass |
| Anti-Bot Fingerprint Rotation | curl-cffi TLS impersonation | Cycles Chrome → Safari → Firefox fingerprints on failure |
| Stealth Scraping | curl-cffi + Playwright | Three-tier bypass: standard HTTP → stealth TLS → full browser automation |
| Streaming Fallback | Telegram Bot API | Sends media URLs directly in chat when download fails |
| Web Search | DuckDuckGo | Real-time web and news search with region/time filters |
Commands:
/create doc <title>|<content> - Word document
/create pdf <title>|<content> - PDF document
/create excel [[...]] - Excel spreadsheet
/create slides [{...}] - PowerPoint
/create image <path> info|thumb|resize|convert|compress|rotate
| Capability | Technology | Description |
|---|---|---|
| Autonomous Coding | Plan → Code → Test → Commit | Full development cycle with iterative test-fix loops and auto-commit |
| Dynamic Tool Registry | PyPI / GitHub hot-install | Discovers, installs, and registers new skills at runtime without restart |
| Capability Marketplace | ClawHub integration | Browse, search, install, update capabilities from clawhub.com |
| Unrestricted System Access | Native OS integrations | Full filesystem, process, and network access with configurable security policies |
| Code Sandboxing | Ephemeral Docker containers | Secure isolated execution for untrusted or generated code |
| Explainability Engine | Decision tracing | Track and explain agent decisions with reasoning chains |
| Capability | Technology | Description |
|---|---|---|
| Multi-Channel Messaging | Discord · Telegram · Slack | Unified ChannelManager routes messages across platforms seamlessly |
| Live Dashboard | FastAPI + WebSocket | Real-time tool log streaming, session history, and interactive chat UI |
| MCP Integration | Model Context Protocol | Native plugin system for GitHub, databases, and third-party tools |
| Visuomotor Automation | PyAutoGUI + MSS | Screen capture and desktop UI interaction for visual automation tasks |
| Model Type | Purpose | Configuration |
|---|---|---|
| Agent Model | General-purpose multimodal | LYA_LLM_DEFAULT_MODEL |
| Vision Model | Image analysis | LYA_LLM_VISION_MODEL |
| Coding Model | Code generation & review | LYA_LLM_CODE_MODEL |
| Reasoning Model | Complex reasoning & architecture | LYA_LLM_REASONING_MODEL |
Lya is configured via environment variables (.env file). The installer walks you through this interactively, or you can edit .env directly:
| Variable | Default | Description |
|---|---|---|
LYA_LLM_PROVIDER |
ollama |
LLM backend: ollama |
LYA_LLM_DEFAULT_MODEL |
llama3 |
Primary multimodal model |
LYA_LLM_VISION_MODEL |
llava |
Image analysis model |
LYA_LLM_CODE_MODEL |
qwen2.5-coder |
Code generation model |
LYA_LLM_REASONING_MODEL |
deepseek-v3 |
Complex reasoning model |
LYA_LLM_TOOL_CALLING_MODE |
auto |
Tool calling: auto, native, prompt |
LYA_PERSONALITY_ENABLED |
true |
Enable personality and emotional state engine |
LYA_PERSONALITY_DEFAULT_TONE |
friendly |
Default tone: friendly, formal, playful, calm |
LYA_TELEGRAM_BOT_TOKEN |
— | Telegram bot token for deployment |
LYA_WORKSPACE_DIR |
~/.lya-workspace |
Directory for downloads, uploads, and projects |
LYA_VOICE_ENABLED |
false |
Enable voice features (setup during installation) |
LYA_SANDBOX_ENABLED |
true |
Enable isolated code execution |
See .env.example for the full configuration reference.
| Section | Description |
|---|---|
| Installation Guide | Detailed setup instructions and requirements |
| Configuration Reference | All environment variables and their effects |
| Architecture Overview | Clean Architecture layers and design decisions |
| Data Flow & CQRS | Command/query separation and event flow |
| Tool & Plugin System | MCP integration and custom tool development |
| Autonomous Governance | Self-Healing, Auto-Updates, and Evolution loops |
| Development Guide | Contributing, testing, and code style |
- Python 3.11 – 3.13
- Git (for installation)
- Ollama (local or remote access)
- Optional: Docker (for code sandboxing), ChromaDB/Qdrant (for persistent memory)
| Component | Requirement |
|---|---|
| CPU | 4+ Physical Cores |
| RAM | 8GB (16GB+ recommended for local LLM) |
| GPU | NVIDIA/AMD with 4GB+ VRAM (highly recommended) |
| Storage | 10GB+ free space (SSD preferred) |
| OS | Linux (Ubuntu/Debian tested), macOS |
Lya is designed to be extended. We welcome contributions of all kinds — new tools, channel integrations, memory backends, and agent capabilities.
- Fork the repository
- Create a feature branch (
git checkout -b feat/my-feature) - Commit your changes (
git commit -m 'feat: add my feature') - Push and open a Pull Request
Please review the Development Guide for code style and testing conventions.
Licensed under the Apache License 2.0.
Copyright 2024–2026 Shojaeei. All rights reserved.