Skip to content

Commit a0cce13

Browse files
phodalphodal
committed
feat(docs): enhance quick start guide with detailed installation and configuration instructions
1 parent bcb27e9 commit a0cce13

File tree

1 file changed

+100
-47
lines changed

1 file changed

+100
-47
lines changed

‎docs/quick-start.md‎

Lines changed: 100 additions & 47 deletions
Original file line numberDiff line numberDiff line change
@@ -6,66 +6,119 @@ description: Get started with AutoDev in minutes
66
import Tabs from '@theme/Tabs';
77
import TabItem from '@theme/TabItem';
88

9-
# Quick Start
10-
11-
This guide will help you get started with AutoDev quickly.
12-
13-
## Installation
9+
# Installation and Setup
1410

1511
<Tabs>
16-
<TabItem value="intellij" label="IntelliJ IDEA" default>
17-
You can install AutoDev from the JetBrains Marketplace:
12+
<TabItem value="marketplace" label="JetBrains Marketplace" default>
13+
Install directly from JetBrains Marketplace Repository: [AutoDev Sketch](https://plugins.jetbrains.com/plugin/26988-autodev-sketch)
1814

19-
1. Open IntelliJ IDEA
20-
2. Go to `Settings` > `Plugins`
21-
3. Search for "AutoDev"
22-
4. Click `Install`
23-
5. Restart IntelliJ IDEA
24-
</TabItem>
25-
<TabItem value="vscode" label="VSCode">
26-
You can install AutoDev from the VSCode Marketplace:
15+
**Note:** The plugin is named `AutoDev Sketch` in the JetBrains Marketplace, and the ID is `com.unitmesh.autodev`. It's different from the one in the custom repository and GitHub releases.
2716

28-
1. Open VSCode
29-
2. Go to the Extensions view
30-
3. Search for "AutoDev"
31-
4. Click `Install`
32-
5. Reload VSCode
17+
This version works for IDE versions 2024.1 and newer (241+).
18+
</TabItem>
19+
<TabItem value="repository" label="Custom Repository">
20+
1. Go to `Settings` → `Plugins` → `Marketplace` → `Manage Plugin Repositories`
21+
2. Add the following URL:
22+
```
23+
https://plugin.unitmesh.cc/updatePlugins.xml
24+
```
25+
</TabItem>
26+
<TabItem value="github" label="GitHub Releases">
27+
1. Download the appropriate version from [GitHub Releases](https://github.com/unit-mesh/auto-dev/releases)
28+
- AutoDev-*-222.zip — For versions 2022.2 to 2023.2
29+
- AutoDev-*-233.zip — For version 2023.3
30+
- AutoDev-*-241.zip — For version 2024.1 and newer
31+
2. Install the plugin from disk in the JetBrains IDE
3332
</TabItem>
3433
</Tabs>
3534

3635
## Configuration
3736

38-
After installation, you need to configure AutoDev:
37+
After installation, configure the plugin in `Settings` → `Tools` → `AutoDev`
3938

40-
1. Go to `Settings` > `Tools` > `AutoDev`
41-
2. Configure your LLM server settings
42-
3. (Optional) Configure custom prompts and agents
39+
### Default LLM
4340

44-
## Basic Usage
41+
Supported providers: Deepseek ([#96](https://github.com/unit-mesh/auto-dev/issues/96)), Moonshot AI, ChatGLM([#90](https://github.com/unit-mesh/auto-dev/issues/960))
4542

46-
### Code Generation
47-
48-
1. Create a new file or open an existing one
49-
2. Use the context menu or keyboard shortcut to access AutoDev features
50-
3. Select the type of code you want to generate
51-
4. Follow the prompts to complete the generation
52-
53-
### Chat with AI
54-
55-
1. Select the code you want to discuss
56-
2. Use the context menu or keyboard shortcut to open the chat
57-
3. Type your question or request
58-
4. Get AI-powered assistance
43+
<Tabs>
44+
<TabItem value="basic" label="Basic Configuration" default>
45+
1. Open AutoDev Config in `Settings` → `Tools` → `AutoDev`
46+
2. Configure `LLM Server Address`, examples:
47+
- Deepseek: `https://api.deepseek.com/chat/completions`
48+
- OpenAI: `https://api.openai.com/v1/chat/completions`
49+
3. Enter your `LLM Key` (API Key)
50+
4. Set `Custom Response Format` using [JsonPath](https://github.com/json-path/JsonPath), example:
51+
- `$.choices[0].delta.content`
52+
5. Configure `Custom Request Format`, example:
53+
- `{ "customFields": {"model": "deepseek-chat", "stream": true }}`
54+
</TabItem>
55+
<TabItem value="advanced" label="Advanced Configuration">
56+
For more detailed configuration options, see [Customize LLM Server](/custom/llm-server)
57+
</TabItem>
58+
</Tabs>
5959

60-
### Auto Testing
60+
### Additional Model
6161

62-
1. Open a test file or create a new one
63-
2. Use the context menu or keyboard shortcut to generate tests
64-
3. Review and modify the generated tests as needed
65-
4. Run the tests to verify functionality
62+
Available model types:
63+
- `Plan`: For reasoning and planning (recommended: DeepSeek R1)
64+
- ~~Act~~: (Not Ready yet) For action execution (e.g., DeepSeek V3, Qwen 72B)
65+
- `Completion`: For code completion
66+
- ~~Embedding~~: (Not Ready yet) For embedding functions (e.g., sentence-transformers/all-MiniLM-L6-v2)
67+
- `FastApply`: For fix patch generation (e.g., Kortix/FastApply-1.5B-v1.0)
68+
- `Others`: Generic placeholder
6669

67-
## Next Steps
70+
<Tabs>
71+
<TabItem value="deepseek" label="DeepSeek R1" default>
72+
```json
73+
{
74+
"name": "DeepSeek R1",
75+
"url": "https://api.deepseek.com/chat/completions",
76+
"auth": {
77+
"type": "Bearer",
78+
"token": "sk-ii"
79+
},
80+
"requestFormat": "{ \"customFields\": {\"model\": \"deepseek-reasoner\", \"stream\": true}}",
81+
"responseFormat": "$.choices[0].delta.content",
82+
"modelType": "Plan"
83+
}
84+
```
85+
</TabItem>
86+
<TabItem value="glm" label="GLM4-Plus">
87+
```json
88+
{
89+
"name": "GLM4-Plus",
90+
"url": "https://open.bigmodel.cn/api/paas/v4/chat/completions",
91+
"auth": {
92+
"type": "Bearer",
93+
"token": "sk-ii"
94+
},
95+
"requestFormat": "{ \"customFields\": {\"model\": \"glm-4-plus\", \"stream\": true}}",
96+
"responseFormat": "$.choices[0].delta.content",
97+
"modelType": "FastApply"
98+
}
99+
```
100+
</TabItem>
101+
<TabItem value="dify" label="DifyAI">
102+
```json
103+
{
104+
"name": "DifyAI",
105+
"description": "Dify Example",
106+
"url": "https://api.dify.ai/v1/completion-messages",
107+
"auth": {
108+
"type": "Bearer",
109+
"token": "app-abcd"
110+
},
111+
"requestFormat": "{\"fields\": {\"inputs\": {\"feature\": \"$content\"}, \"response_mode\": \"streaming\", \"user\": \"phodal\" }}",
112+
"responseFormat": "$.answer",
113+
"modelType": "Others"
114+
}
115+
```
116+
</TabItem>
117+
</Tabs>
68118

69-
- Learn about [Features](/docs/features)
70-
- Explore [Customization](/docs/customize)
71-
- Check out [Troubleshooting](/docs/troubleshooting) if you encounter issues
119+
Configuration fields:
120+
- **URL**: LLM Server Address with endpoint path
121+
- **Auth**: Authentication information (currently Bearer token only)
122+
- **RequestFormat**: JSON structure for API requests
123+
- **ResponseFormat**: JsonPath to extract content from responses
124+
- **ModelType**: Type of model (see list above)

0 commit comments

Comments
 (0)