Loading repository data…
Loading repository data…
linhai0872 / repository
Production-grade AI Agent orchestration scaffold based on Agno + AgentOS. Supports 8 model providers (OpenRouter/OpenAI/Google/Anthropic/Alibaba/Volcengine), multi-agent Teams, and conditional Workflows.
A transparent discovery signal based on current public GitHub metadata.
This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.
Production-grade AI Agent scaffold with batteries included
Build, deploy, and scale intelligent agents in minutes, not weeks.
📖 Docs · 🍳 Cookbook · Issues
docker compose upModelConfig for all providers, switch with one line# Create a new project
npx create-agno-agent my-agent
cd my-agent
# Configure & Start
cp .env.example .env # Add OPENROUTER_API_KEY
docker compose up -d
# Done! Open http://localhost:7777/docs
git clone https://github.com/linhai0872/agno-agent-starter.git && cd agno-agent-starter
cp .env.example .env
docker compose up -d
from app.agents.github_analyzer import create_github_analyzer_agent
from agno.db.postgres import PostgresDb
db = PostgresDb(db_url="postgresql://...")
agent = create_github_analyzer_agent(db)
response = agent.run("Analyze https://github.com/agno-agi/agno")
# Returns: GitHubRepoAnalysis (structured output)
docker compose -f docker-compose.yml -f docker-compose.prod.yml up -d
http://localhost:7777 → ConnectModelConfig, no hardcoding┌─────────────────────────────────────────────────────────────────┐
│ AgentOS Runtime │
├─────────────────────────────────────────────────────────────────┤
│ Agents │ Teams │ Workflows │
│ Single task │ Multi-agent │ Step-based flow │
├─────────────────────────────────────────────────────────────────┤
│ Core Abstractions │
│ Models │ Tools │ Hooks │ Registry │
│ Unified │ 3-tier Reg │ Guardrails │ PriorityRegistry │
├─────────────────────────────────────────────────────────────────┤
│ Infrastructure │
│ PostgreSQL + pgvector + Tracing │
└─────────────────────────────────────────────────────────────────┘
| Template | Type | Description | Key Features |
|---|---|---|---|
| GitHub Analyzer | Agent | Analyze GitHub repos | Structured Output, DuckDuckGo |
| Deep Research | Team | Multi-agent research | session_state, ToolCallGuard |
| Customer Service | Workflow | Smart support with RAG | Condition routing, pgvector |
| Tool | For | Description |
|---|---|---|
| Agno CLI | Developers | Terminal tool for running and debugging Agent/Team/Workflow |
| MCP DevTools | AI Agents | MCP tools for AI to test and debug Agno apps (zero-config) |
# CLI: Run a workflow and auto-trace
./scripts/agno run workflow customer-service '{"query":"help"}'
# CLI: System health check
./scripts/agno health
agno-agent-starter/
├── app/
│ ├── main.py # AgentOS entrypoint
│ ├── config.py # 3-tier config loader
│ ├── agents/ # ✏️ Agent implementations
│ ├── teams/ # ✏️ Team implementations
│ ├── workflows/ # ✏️ Workflow implementations
│ ├── models/ # Model abstraction (8 providers)
│ ├── tools/ # 3-tier tool registry
│ ├── hooks/ # Guardrails & lifecycle hooks
│ ├── mcp/ # MCP Server (AI Agent tools)
│ └── core/ # Registry abstractions
├── api/ # OpenAPI spec (auto-generated)
├── scripts/ # CLI tools (agno)
├── tests/ # Unit tests
└── .cursor/rules/ # Vibe Coding rules
✏️ = User extension points
| Mode | Use Case | Description |
|---|---|---|
| Agent | Single task + Tools | 90% recommended, simple & efficient |
| Team | Multi-role collaboration | Auto-coordination between members |
| Workflow | Strict steps + Conditions | Process control & state management |
| Variable | Required | Description |
|---|---|---|
OPENROUTER_API_KEY | Yes | OpenRouter API Key |
DATABASE_URL | No | PostgreSQL (default from Docker) |
DEBUG_MODE | No | Hot reload for dev |
Full config: .env.example
# Export latest spec
python scripts/export_openapi.py
Issues and PRs are welcome! Please read AGENTS.md for development guidelines.