Sniping bargains on Xianyu can now be left to AI watching for you 24/7! 🔗 : https://t.co/t2kWLa9PX0 This tool is called ai-goofish-monitor — essentially an automated Xianyu deal-hunting bot. It monitors all listings in real time, uses AI to judge value, authenticity, and pricing reasonableness, and instantly pushes notifications for unusually low prices. Still manually scrolling through thousands of posts? You’re already losing. https://t.co/riPAin91Z8

Usagi-org/ai-goofish-monitor

Source: https://github.com/Usagi-org/ai-goofish-monitor

Xianyu Smart Monitoring System

[中文] | English

A real-time multi‑task Xianyu monitor powered by Playwright and AI, with a full Web management interface.

Core Features

• Web‑based Visual Management: task management, account management, AI prompt editing, run logs, result browsing.
• AI‑Driven: create tasks in natural language, deep product analysis with multimodal models.
• Multi‑Task Concurrency: independent configuration of keywords, prices, filters, and AI prompts.
• Advanced Filtering: free shipping, new listing time range, province/city/district three‑level region filter.
• Instant Notifications: support for ntfy.sh, WeCom, Bark, Telegram, Webhook, and more.
• Scheduled Execution: Cron configuration for periodic tasks.
• Account & Proxy Rotation: multi‑account management, task‑account binding, proxy pool rotation with failure retry.
• Docker Deployment: one‑click containerized setup.

Screenshots

Monitoring Overview Task Management Result Viewing Notification Push

🐳 Docker Deployment (Recommended)

git clone https://github.com/Usagi-org/ai-goofish-monitor && cd ai-goofish-monitor
cp .env.example .env
vim .env            # Fill in relevant configuration
docker compose up -d
docker compose logs -f app
docker compose down

If the image cannot be accessed or the download speed is slow, you can try using an accelerator:

docker pull ghcr.nju.edu.cn/usagi-org/ai-goofish:latest
docker tag ghcr.nju.edu.cn/usagi-org/ai-goofish:latest ghcr.io/usagi-org/ai-goofish:latest
docker compose up -d

• Default Web UI address: http://127.0.0.1:8000
• The Docker image already includes Chromium, no additional browser installation on the host is required.
• Official image address: ghcr.io/usagi-org/ai-goofish:latest
• Update the image: docker compose pull && docker compose up -d
• If you modify SERVER_PORT in .env, update the port mapping in docker-compose.yaml accordingly.
• docker-compose.yaml mounts the SQLite main database to ./data:/app/data by default; the database file is data/app.sqlite3.
• The following directories are persisted by default:
  • data/ SQLite main storage (tasks, results, price history)
  • state/ login state cookie files
  • prompts/ task prompts
  • logs/ run logs
  • images/ product images and task temporary images
  • config.json, jsonl/, price_history/ – legacy data sources that are imported on first upgrade to SQLite.

Data Storage & Migration

• The current primary storage is SQLite, default path data/app.sqlite3.
• The database path can be customized via the environment variable APP_DATABASE_FILE; in Docker it defaults to /app/data/app.sqlite3.
• On startup, the application automatically creates the database and tables, and attempts to import historical data from legacy config.json, jsonl/, and price_history/ once.
• state/, prompts/, logs/, images/ remain file‑system directories and are not stored in SQLite.
• Product images are temporarily saved to images/task_images_<task_id>/ and are cleaned up by default after the task finishes.
• After the initial upgrade and once you confirm the data in data/app.sqlite3 is correct, you may decide whether to keep the legacy mounts (config.json, jsonl/, price_history/) depending on your deployment approach.

Minimum Configuration

Se e the “Configuration Guide” section below for other settings.

First‑Time Use

1. Open the default Web UI at http://127.0.0.1:8000 and log in.
2. Go to “Xianyu Account Management”, use the Chrome extension (https://chromewebstore.google.com/detail/xianyu-login-state-extrac/eidlpfjiodpigmfcahkmlenhppfklcoa) to export and paste a Xianyu login state JSON.
3. The login state file will be saved to the state/ directory, e.g., state/acc_1.json.
4. Return to “Task Management”, create a task, bind an account, and run it.

Creating Your First Task

• AI Judgment: Fill in “Detailed Requirements” and submit; a separate progress popup appears, and the analysis standard is generated asynchronously.
• Keyword Judgment: Fill in keyword rules; the task is created immediately without going through the AI generation process.
• Region Filter: Changed to a three‑level selector (province/city/district), with data built‑in from page snapshots.

User Guide

Developer Guide

Requirements

• Python 3.10+
• Node.js + npm (verified locally with Node v20.18.3 can complete frontend build)
• Playwright CLI and Chromium. It is recommended to run python3 -m pip install playwright && python3 -m playwright install chromium before the first run.
• Chrome / Edge browser (Chromium works in Linux; start.sh checks for browser availability first)

git clone https://github.com/Usagi-org/ai-goofish-monitor
cd ai-goofish-monitor
cp .env.example .env

One‑Click Start

chmod +x start.sh
./start.sh

start.sh first checks Playwright CLI and browser prerequisites; once met, it automatically installs project dependencies, builds the frontend, copies the build artifacts, and starts the backend.

Manual Start

# Backend
python -m src.app
# Or
uvicorn src.app:app --host 0.0.0.0 --port 8000 --reload

# Frontend
cd web-ui
npm install
npm run dev

• FastAPI on startup automatically initializes SQLite and attempts to import legacy config.json / jsonl / price_history on the first run.
• spider_v2.py reads tasks from SQLite by default; only when --config <path> is explicitly passed does it fall back to the JSON compatibility mode.
• Default database path: data/app.sqlite3.
• The Vite dev server proxies /api, /auth, /ws to http://127.0.0.1:8000.
• npm run build generates web-ui/dist/; start.sh then copies it to the repo root dist/.
• FastAPI serves root dist/index.html and dist/assets/.
• ./start.sh outputs the access URL http://localhost:8000 and API documentation http://localhost:8000/docs.

Tests & Validation

PYTEST_DISABLE_PLUGIN_AUTOLOAD=1 pytest
cd web-ui && npm run build

Task Creation API

Configuration Guide

Web Interface Authentication

🚀 Workflow

The diagram below shows the core processing logic of a single monitoring task from start to finish. The main service runs in src.app and launches one or more task processes based on user actions or scheduled triggers.

graph TD
    A[Start monitoring task] --> B[Select account/proxy configuration];
    B --> C[Task: search products];
    C --> D{New product found?};
    D -- Yes --> E[Fetch product details & seller info];
    E --> F[Download product images];
    F --> G[Call AI for analysis];
    G --> H{AI recommends?};
    H -- Yes --> I[Send notification];
    H -- No --> J[Save record to SQLite];
    I --> J;
    D -- No --> K[Next page / wait];
    K --> C;
    J --> C;
    C --> L{Anti‑spam / exception triggered?};
    L -- Yes --> M[Rotate account/proxy and retry];
    M --> C;

Frequently Asked Questions

Acknowledgements

Notice

Star History

Star History Chart