y-66/traveler

Эта страница переведена автоматически. Технические детали смотрите в оригинале. Открыть оригинал

20 stars 1 forks Качество 49 Тренд 10

An enterprise-grade travel planning intelligent agent system built by Agno framework. Integrating advanced functionalities including MCP, Skills, Memory, Knowledge Bases, Agent Teams, and Workflows.

Обзор

is an enterprise-grade travel planning intelligent agent system built entirely upon the Agno framework. By integrating advanced functionalities including , , , , , and , Traveler provides highly customized, comprehensive, and up-to-date travel planning services. The system includes a rich CLI powered by Typer. Traveler can run as a stateful FastAPI backend. To empower the agent with your own customized guides: 1. Place PDF/Markdown guides in src/traveler/knowledge_docs/ 2. Run the load command to vectorize them into local ChromaDB: - : Core Agentic Framework (Agent, Team, Workflow, MCP, Skills) - : Robust Web API & Data Validation - : Vector Databases for Knowledge (RAG) - : Session Memory & Persistence Base (Easily swapped to Postgres) - : Powerful and beautiful CLI creation - : Modern Python logging 是一个基于 Agno 框架构建的企业级旅游规划智能体系统。通过深度集成、、、、以及等能力，Traveler 为用户提供高度定制化、全面且信息实时的旅行规划服务。如果您有特定的旅游攻略或私有数据： 1. 请将 Markdown 或 PDF 文档放入 src/traveler/knowledge_docs/ 目录下。 2.

README

Показан фрагмент оригинального README. Открыть оригинал.

🌍 Agno-Based Travel Planning Agent System / 基于Agno的旅游规划智能体

English Version

Traveler is an enterprise-grade travel planning intelligent agent system built entirely upon the Agno framework. By integrating advanced functionalities including MCP (Model Context Protocol), Skills, Memory, Knowledge Bases, Agent Teams, and Workflows, Traveler provides highly customized, comprehensive, and up-to-date travel planning services.

Result:

English: Report presentaion
Chinese: 报告展示

✨ Core Features

Feature	Description
Agent	Travel planner agent based on Agno, capable of complex tool calling, multi-turn dialogue, and autonomous reasoning.
Team Collaboration	Multi-agent collaboration model: orchestrates a Planner, Researcher, and Budget Analyst to work together seamlessly.
Workflow	Structured pipeline execution: Research (Information Gathering) → Budgeting → Itinerary Planning.
MCP Integration	Connects to external MCP servers to fetch real-time Weather, generic Web Search, Hotel integrations, and Google Route analysis.
Skills plugins	Specialized skill modules (Agno Skills) for Travel Planning, Local Expertise, Intent Analysis, and Budget Optimization.
Memory System	Automatically retains and retrieves user preferences and interaction history for fully personalized sessions.
Knowledge Base	RAG (Retrieval-Augmented Generation) functionality supported by ChromaDB/LanceDB locally to store and query travel guide documents.
Custom Tools	Built-in toolkit tailored for flight, hotel, and attraction search scenarios.

🏗️ Architecture & Structure

Traveler/
├── .env.example                # Environment variables template
├── pyproject.toml              # Project configuration and dependencies (Hatch build system)
├── data/                       # Local data storage (SQLite, ChromaDB outputs, reports)
├── src/traveler/               # Main source code directory
│   ├── agents/                 # Agent Definitions (Planner, Researcher, Budget Analyst, Team)
│   ├── api/                    # FastAPI web service (app, routes, schemas)
│   ├── config/                 # Pydantic Settings & Loguru configurations
│   ├── core/                   # Core Infrastructure (Model Factory, Database Setup, Knowledge Management)
│   │   └── mcp/                # External MCP Server Clients (Weather, Google Routes, etc.)
│   ├── knowledge_docs/         # Document drop-zone for RAG Knowledge base

Открыть полный README на GitHub