Quick Start

This guide helps you go from registration to your first conversation.

---

In This Guide

• Register & Login
• First-time Setup
• Dashboard Overview
• API Key Management
• Create & Configure Agent
• Start Chatting

---

1. Register & Login

1. Visit https://app.aigent.ren
2. Register with email or third-party login (Google/GitHub)
3. After email verification, login with your email and password

---

2. First-time Setup

After logging in, complete the setup in this order:

Step 1: Add API Key → Step 2: Create Agent → Step 3: Start Chatting

Step	Action	Entry Point
1. Add API Key	Configure LLM provider's API Key	Settings → API Keys
2. Create Agent	Create your first AI assistant	Dashboard → Create Agent
3. Start Chatting	Click Agent card to enter chat	Dashboard → Agent list

⚠️ Important: You must add an API Key first, otherwise Agent cannot run.

---

3. Dashboard Overview

Dashboard is the homepage after login, containing these areas:

3.1 Quick Stats

Top of page shows:

• Number of Agents
• Recent tasks
• Number of API Keys

3.2 Recent Tasks

Middle of page shows recent task list for all Agents, for quick overview of activity.

3.3 Agent List

Bottom of page shows all your Agents:

• Click Agent card → Enter chat page
• Click settings icon on card → Enter Agent settings
• Click delete → Delete the Agent

3.4 Welcome Guide (New Users)

If you're logging in for the first time with no Agents, a welcome guide appears:

• Shows quick start steps
• Prompts to add API Key first
• Provides "Create Agent" shortcut

---

4. API Key Management

4.1 Why Do I Need an API Key?

Agents need to call LLM (Large Language Models) to respond to your requests. You need to provide the corresponding provider's API Key to use them.

4.2 How to Get an API Key?

Provider	Get URL
Anthropic	https://console.anthropic.com
OpenAI	https://platform.openai.com/api-keys
Google (Gemini)	https://aistudio.google.com/apikey
DeepSeek	https://platform.deepseek.com
Qwen	https://dashscope.console.aliyun.com
Zhipu	https://open.bigmodel.cn

4.3 Add API Key

1. Click "Settings" → "API Keys" in left sidebar
2. Click "Add Key" button
3. Select provider, enter API Key, add label (e.g., "for work")
4. Click save, Key will be encrypted and stored

4.4 Security Notes

• API Keys are encrypted with Fernet
• Never transmitted or displayed in plain text
• Can be added/deleted anytime

---

5. Create & Configure Agent

5.1 Create Agent

1. Click "Create Agent" button on Dashboard
2. Fill in configuration:
   • Name: Agent display name (required)
   • LLM Provider: Select AI model provider
   • Model: Select specific model
   • API Key: Select API Key to use
   • Skills: Bind predefined skills (optional)
   • Risk Level: Confirmation level for tool calls
3. Click "Create Agent" to complete

5.2 Available LLM Providers

Provider	Model Examples	Features
Anthropic	Claude 3.5/3.7/4 series	Strong reasoning
OpenAI	GPT-4o/4.1/4.5	Good versatility
Gemini	Gemini 1.5/2.0/2.5	Multimodal support
DeepSeek	DeepSeek-V3/DeepSeek-Reasoner	Cost-effective
Qwen	Qwen-Turbo/Plus	Chinese optimized
Zhipu	GLM-4/Plus	Domestic LLM

5.3 Risk Level Explanation

Level	Behavior	Use Case
low	Auto-execute all tools	Trusted tool environment
medium	Medium+ risk requires confirmation	Daily use (recommended)
high	All tools require confirmation	Cautious operation scenarios

Auto Confirm Low Risk: After enabling "Auto Confirm Low Risk", low-risk tools will execute automatically without manual confirmation.

5.4 Modify Agent Configuration

1. Find Agent card on Dashboard
2. Click settings icon (gear) on card
3. Modify configuration and click "Save Changes"

---

6. Start Chatting

6.1 Enter Chat Page

Click Agent card body (not settings icon) on Dashboard to enter chat page.

6.2 Chat Interface Layout

┌─────────────────────────────────────────────────────┐
│ Top bar: Agent Name | Memory | Settings             │
├─────────────────────────────────┬───────────────────┤
│                                 │                   │
│     Message area                │   Task history    │
│     (conversation display)      │   sidebar         │
│                                 │                   │
├─────────────────────────────────┴───────────────────┤
│ Input box + Send/Stop buttons                       │
└─────────────────────────────────────────────────────┘

6.3 Send Message

• Enter content in bottom input box
• Press Enter or click "Send" to send
• AI will stream response in real-time

6.4 Stop Generation

To interrupt AI response:

• Click red "Stop" button on right side of input box
• AI will stop generating, output content is preserved

6.5 Tool Call Confirmation

When Agent wants to call a higher-risk tool:

1. Conversation pauses, confirmation popup appears
2. Shows tool call details
3. Choose "Confirm" to allow or "Cancel" to reject

---

Next Steps

Congratulations on completing your first conversation! Next you can:

• 📖 Feature Guide - Learn about skills system, memory management, and other advanced features
• ❓ Help & Support - Having issues? Check the FAQ