# aiagent

**Repository Path**: hali-java/aiagent

## Basic Information

- **Project Name**: aiagent
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-04-07
- **Last Updated**: 2026-04-08

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Student Exam AI

Exam upload, error analysis, and remediation paper generation SaaS scaffold.

## Structure

```text
student-exam-ai/
  api/         Go API, PostgreSQL persistence, SaaS data model
  ml/          Python ML service, now using LangChain + OpenAI-compatible chat models
  web/         React frontend for tenant / school / class scoped workflows
  scripts/     Local start script
```

## Current state

- Go API persists tenants, schools, classes, students, reports, and practice papers
- Python ML service supports LangChain-based structured output generation
- React frontend scopes intake and history by tenant, school, and class
- React frontend now includes an analysis chat workspace for exam and class follow-up questions

## AI configuration

The ML service is now wired for Alibaba Bailian through the DashScope OpenAI-compatible endpoint.

```powershell
cd ml
copy .env.example .env
```

Main variables:

- `DASHSCOPE_API_KEY`
- `BAILIAN_BASE_URL`
- `BAILIAN_MODEL`
- `LLM_TEMPERATURE`
- `AI_USE_MOCK_FALLBACK`
- `WEB_SEARCH_ENABLED`
- `WEB_SEARCH_TIMEOUT_SECONDS`

## Local startup

### 1. Install Python dependencies

```powershell
cd ml
python -m venv .venv
.\.venv\Scripts\python -m pip install -r requirements.txt
```

### 2. Start Python ML service

```powershell
cd ml
.\.venv\Scripts\python src/app.py
```

### 3. Start Go API

```powershell
cd api
go run ./cmd/server
```

### 4. Start React frontend

```powershell
cd web
npm install
npm run dev
```

Default ports:

- Python ML service: `8001`
- Go API: `8080`
- React frontend: `5173`

## PowerShell startup scripts

- `ml/run.ps1`：启动 Python ML 服务
- `api/run.ps1`：启动 Go API 服务
- `web/run.ps1`：启动 React 前端
- `scripts/start-stack.ps1`：按顺序启动三层服务
- `scripts/check-stack.ps1`：检查三层健康状态
- `scripts/stop-stack.ps1`：停止由总控脚本启动的服务

Recommended flow:

```powershell
.\scripts\start-stack.ps1
.\scripts\check-stack.ps1
```