Skip to content

Commit 067106a

Browse files
Merge pull request volcengine#28 from volcengine/feat/project-info
feat: add project readme
2 parents 1bd73cf + 7770151 commit 067106a

File tree

11 files changed

+344
-17
lines changed

11 files changed

+344
-17
lines changed

02-use-cases/ai_coding/README.md

Lines changed: 22 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -44,18 +44,26 @@ AI 编程助手
4444

4545
1. 登录 [火山引擎控制台](https://console.volcengine.com)
4646
2. 进入"访问控制" → "用户" -> 新建用户 或 搜索已有用户名 -> 点击用户名进入"用户详情" -> 进入"密钥" -> 新建密钥 或 复制已有的 AK/SK
47+
- 如下图所示
48+
![Volcengine AK/SK Management](../img/volcengine_aksk.jpg)
4749
3. 为用户配置 AgentKit运行所依赖服务的访问权限:
4850
- 在"用户详情"页面 -> 进入"权限" -> 点击"添加权限",将以下策略授权给用户
4951
- `AgentKitFullAccess`(AgentKit 全量权限)
5052
- `APMPlusServerFullAccess`(APMPlus 全量权限)
5153
4. 为用户获取火山方舟模型 Agent API Key
52-
- 搜索"火山方舟"产品,点击进入控制台
53-
- 进入"API Key管理" -> 创建 或 复制已有的 API Key
54+
- 登陆[火山方舟控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/overview?briefPage=0&briefType=introduce&type=new)
55+
- 进入"API Key管理" -> 创建 或 复制已有的 API Key,后续`MODEL_AGENT_API_KEY`环境变量需要配置为该值
56+
- 如下图所示
57+
![Ark API Key Management](../img/ark_api_key_management.jpg)
5458
5. 开通模型预置推理接入点
55-
- 搜索"火山方舟"产品,点击进入控制台
59+
- 登陆[火山方舟控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/overview?briefPage=0&briefType=introduce&type=new)
5660
- 进入"开通管理" -> "语言模型" -> 找到相应模型 -> 点击"开通服务"
5761
- 确认开通,等待服务生效(通常1-2分钟)
58-
- 本案例使用的是`deepseek-v3-1-terminus`模型,因此需要开通`deepseek-v3-1-terminus`模型的预置推理接入点(您也可以根据实际需求开通其他模型的预置推理接入点,并在`agent.py`代码中指定使用的模型)
62+
- 开通本案例中使用到的以下模型
63+
- `deepseek-v3-1-terminus`
64+
- `doubao-seed-code-preview-251028`
65+
- 如下图所示
66+
![Ark Model Service Management](../img/ark_model_service_management.jpg)
5967

6068
**2. AgentKit 工具 ID**
6169

@@ -64,7 +72,8 @@ AI 编程助手
6472
3. 创建工具:
6573
- 工具名称: `ai-coding-agent`
6674
- 描述: AI 编程助手工具
67-
4. 复制生成的工具 ID 用于配置
75+
4. 复制生成的工具 ID 用于配置(后续`AGENTKIT_TOOL_ID`环境变量需要配置为该值), 如下图所示
76+
![AgentKit Sandbox Tool](../img/agentkit_sandbox_tool.jpg)
6877

6978
### 安装依赖
7079

@@ -99,7 +108,14 @@ export MODEL_AGENT_API_KEY={{your_model_agent_api_key}} # 从火山方舟获取
99108
```
100109

101110
**环境变量说明:**
102-
- `DATABASE_TOS_BUCKET`: 用于存储生成的前端代码的 TOS 存储桶
111+
- `VOLCENGINE_ACCESS_KEY`: 火山引擎访问凭证的 Access Key
112+
- `VOLCENGINE_SECRET_KEY`: 火山引擎访问凭证的 Secret Key
113+
- `DATABASE_TOS_BUCKET`: 用于存储生成的前端代码的 TOS 存储桶名称
114+
- 格式: `DATABASE_TOS_BUCKET=agentkit-platform-{{your_account_id}}`
115+
- 示例: `DATABASE_TOS_BUCKET=agentkit-platform-12345678901234567890`
116+
- 其中`{{your_account_id}}`需要替换为您的火山引擎账号 ID
117+
- `AGENTKIT_TOOL_ID`: 从 AgentKit 控制台获取的工具 ID
118+
- `MODEL_AGENT_API_KEY`: 从火山方舟获取的模型 Agent API Key
103119

104120
## 本地测试
105121

File renamed without changes.

02-use-cases/customer_support/README.md

Lines changed: 28 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -49,16 +49,31 @@ AgentKit 运行时
4949
**Python 版本**
5050
- 需要 Python 3.12 或更高版本
5151

52-
**火山引擎服务**
53-
1. **访问凭证**: 登录 [火山引擎控制台](https://console.volcengine.com)
54-
- 进入"访问控制" → "密钥管理"
55-
- 点击"创建密钥"生成 Access Key 和 Secret Key
56-
- 为凭证配置以下产品权限:
57-
- AgentKit (智能体运行时和工具)
58-
- TOS (知识库文件存储)
59-
- Viking (知识和记忆的向量数据库)
60-
61-
2. **知识库** (首次运行自动配置):
52+
**1. 火山引擎访问凭证**
53+
54+
1. 登录 [火山引擎控制台](https://console.volcengine.com)
55+
2. 进入"访问控制" → "用户" -> 新建用户 或 搜索已有用户名 -> 点击用户名进入"用户详情" -> 进入"密钥" -> 新建密钥 或 复制已有的 AK/SK
56+
- 如下图所示
57+
![Volcengine AK/SK Management](../img/volcengine_aksk.jpg)
58+
3. 为用户配置 AgentKit运行所依赖服务的访问权限:
59+
- 在"用户详情"页面 -> 进入"权限" -> 点击"添加权限",将以下策略授权给用户
60+
- `AgentKitFullAccess`(AgentKit 全量权限)
61+
- `APMPlusServerFullAccess`(APMPlus 全量权限)
62+
4. 为用户获取火山方舟模型 Agent API Key
63+
- 登陆[火山方舟控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/overview?briefPage=0&briefType=introduce&type=new)
64+
- 进入"API Key管理" -> 创建 或 复制已有的 API Key,后续`MODEL_AGENT_API_KEY`环境变量需要配置为该值
65+
- 如下图所示
66+
![Ark API Key Management](../img/ark_api_key_management.jpg)
67+
5. 开通模型预置推理接入点
68+
- 登陆[火山方舟控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/overview?briefPage=0&briefType=introduce&type=new)
69+
- 进入"开通管理" -> "语言模型" -> 找到相应模型 -> 点击"开通服务"
70+
- 确认开通,等待服务生效(通常1-2分钟)
71+
- 开通本案例中使用到的以下模型(您也可以根据实际需求开通其他模型的预置推理接入点,并在`agent.py`代码中指定使用的模型)
72+
- `deepseek-v3-1-terminus`
73+
- 如下图所示
74+
![Ark Model Service Management](../img/ark_model_service_management.jpg)
75+
76+
**2. 知识库(首次运行自动配置)**:
6277
- 如未设置 `DATABASE_VIKING_COLLECTION`,智能体将自动:
6378
- 上传 `pre_build/knowledge/` 中的文件到 TOS
6479
- 创建 Viking 集合
@@ -108,6 +123,9 @@ export DATABASE_MEM0_API_KEY=<mem0_api_key>
108123

109124
**环境变量说明:**
110125
- `DATABASE_TOS_BUCKET`: 用于自动知识库初始化所需。若未设置 `DATABASE_VIKING_COLLECTION`,首次运行会将 `pre_build/knowledge` 自动上传至 TOS 并导入 Viking 向量库。
126+
- 格式: `DATABASE_TOS_BUCKET=agentkit-platform-{{your_account_id}}`
127+
- 示例: `DATABASE_TOS_BUCKET=agentkit-platform-12345678901234567890`
128+
- 其中`{{your_account_id}}`需要替换为您的火山引擎账号 ID
111129
- `DATABASE_VIKING_COLLECTION`: 预创建的知识库集合名称 (生产环境推荐)
112130
- 模型默认为 `deepseek-v3-1-terminus` ,如需更改可在代码中调整。
113131

487 KB
Loading
609 KB
Loading
1.37 MB
Loading
749 KB
Loading

02-use-cases/video_gen/README.md

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -59,20 +59,26 @@ AgentKit 运行时
5959

6060
1. 登录 [火山引擎控制台](https://console.volcengine.com)
6161
2. 进入"访问控制" → "用户" -> 新建用户 或 搜索已有用户名 -> 点击用户名进入"用户详情" -> 进入"密钥" -> 新建密钥 或 复制已有的 AK/SK
62+
- 如下图所示
63+
![Volcengine AK/SK Management](../img/volcengine_aksk.jpg)
6264
3. 为用户配置 AgentKit运行所依赖服务的访问权限:
6365
- 在"用户详情"页面 -> 进入"权限" -> 点击"添加权限",将以下策略授权给用户
6466
- `AgentKitFullAccess`(AgentKit 全量权限)
6567
- `APMPlusServerFullAccess`(APMPlus 全量权限)
6668
4. 为用户获取火山方舟模型 Agent API Key
6769
- 搜索"火山方舟"产品,点击进入控制台
6870
- 进入"API Key管理" -> 创建 或 复制已有的 API Key
71+
- 如下图所示
72+
![Ark API Key Management](../img/ark_api_key_management.jpg)
6973
5. 开通模型预置推理接入点
7074
- 搜索"火山方舟"产品,点击进入控制台
7175
- 进入"开通管理" -> "语言模型" -> 找到相应模型 -> 点击"开通服务"
7276
- 开通本案例中使用到的以下模型
7377
- root_agent模型:`deepseek-v3-1-terminus`
7478
- 生图模型:`doubao-seedream-4-0-250828`
7579
- 生视频模型:`doubao-seedance-1-0-pro-250528`
80+
- 如下图所示
81+
![Ark Model Service Management](../img/ark_model_service_management.jpg)
7682

7783
### 安装依赖
7884

@@ -112,6 +118,8 @@ export DOWNLOAD_DIR=/tmp
112118

113119
**TOS 存储桶配置:**
114120
- 默认存储桶: `agentkit-platform-{{your_account_id}}`
121+
- 其中`{{your_account_id}}`需要替换为您的火山引擎账号 ID
122+
- 示例: `DATABASE_TOS_BUCKET=agentkit-platform-12345678901234567890`
115123
- 若需自定义,可在 [`tool/tos_upload.py`](tool/tos_upload.py) 中修改 `bucket_name` 参数或在工具调用时传入
116124

117125
## 本地测试

03-integrations/README.md

Whitespace-only changes.

README.md

Lines changed: 143 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1,143 @@
1-
# agentkit-samples
1+
<div align="center">
2+
<h1>
3+
AgentKit Platform Python Samples
4+
</h1>
5+
6+
<div align="center">
7+
<a href="https://github.com/volcengine/agentkit-samples/graphs/commit-activity"><img alt="GitHub commit activity" src="https://img.shields.io/github/commit-activity/m/volcengine/agentkit-samples"/></a>
8+
<a href="https://github.com/volcengine/agentkit-samples/pulls"><img alt="GitHub open pull requests" src="https://img.shields.io/github/issues-pr/volcengine/agentkit-samples"/></a>
9+
<a href="https://github.com/volcengine/agentkit-samples/blob/main/LICENSE"><img alt="License" src="https://img.shields.io/github/license/volcengine/agentkit-samples"/></a>
10+
<a href="https://python.org"><img alt="Python versions" src="https://img.shields.io/pypi/pyversions/agentkit-sdk-python"/></a>
11+
</div>
12+
13+
<p>
14+
<a href="https://console.volcengine.com/agentkit/"> Volcengine AgentKit</a>
15+
◆ <a href="https://volcengine.github.io/agentkit-sdk-python/">SDK/CLI Documentation</a>
16+
◆ <a href="https://github.com/volcengine/agentkit-samples/tree/main">Samples</a>
17+
◆ <a href="https://pypi.org/project/agentkit-sdk-python/">PyPI Package</a>
18+
◆ <a href="https://github.com/volcengine/agentkit-sdk-python">SDK/CLI GitHub</a>
19+
20+
</p>
21+
</div>
22+
23+
# AgentKit Samples
24+
25+
Welcome to the AgentKit Samples repository!
26+
27+
AgentKit is an enterprise-level AI Agent development platform launched by Volcengine, providing developers with complete solutions for Agent construction, deployment, and operation. Through standardized development toolchains and cloud-native infrastructure, the platform significantly lowers the development and deployment threshold for complex intelligent agent applications.
28+
29+
This repository contains a collection of examples and tutorials to help you understand, implement, and integrate AgentKit functionalities into your applications.
30+
31+
## Project Structure
32+
```bash
33+
.
34+
├── 01-tutorials
35+
│ └── README.md
36+
├── 02-use-cases
37+
│ ├── ai_coding
38+
│ ├── beginner
39+
│ │ ├── a2a_simple
40+
│ │ ├── callback
41+
│ │ ├── episode_generation
42+
│ │ ├── hello_world
43+
│ │ ├── mcp_simple
44+
│ │ ├── multi_agents
45+
│ │ ├── restaurant_ordering
46+
│ │ ├── travel_concierge
47+
│ │ ├── vikingdb
48+
│ │ ├── vikingmem
49+
│ │ └── README.md
50+
│ ├── customer_support
51+
│ └── video_gen
52+
├── README.md
53+
└── README.zh.md
54+
```
55+
56+
### 01-tutorials/ - Interactive Learning & Fundamentals (Coming Soon)
57+
58+
This folder will contain tutorial-based learning materials that teach AgentKit's core functionalities through practical examples.
59+
60+
**Component Categories:**
61+
62+
- **Runtime**: AgentKit runtime environment, providing secure and scalable agent deployment capabilities
63+
- **Gateway**: Tool gateway, automatically converting APIs and external services into agent-usable tools
64+
- **Memory**: Agent memory management, supporting cross-session, context-aware, and personalized interactions
65+
- **Identity**: Agent identity authentication and permission control, building security trust mechanisms across the user→Agent→tool chain
66+
- **Tools**: Built-in toolset, including code interpreter and browser tools
67+
- **Observability**: Agent observability, providing tracing, debugging, and monitoring capabilities
68+
69+
These examples are perfect for beginners and users who want to understand core concepts before building actual agent applications.
70+
71+
### 02-use-cases/ - End-to-End Application Examples
72+
73+
Explore practical business scenario implementations that demonstrate how to apply AgentKit functionalities to solve real-world business problems.
74+
75+
**Current Use Cases:**
76+
77+
- **ai_coding/**: AI coding assistant, helping developers write and optimize code
78+
- **beginner/**: Beginner-level examples, from basic to advanced agent development
79+
- **customer_support/**: Customer support agent, providing automated after-sales consulting and pre-sales guidance
80+
- **video_gen/**: Video generation agent, combining multiple tools for video content creation
81+
82+
Each use case includes complete implementations with detailed explanations on how to combine AgentKit components to build applications.
83+
84+
## Quick Start
85+
86+
### Prerequisites
87+
88+
- Python 3.10+
89+
- AgentKit SDK
90+
- Optional: Docker (for containerized deployment)
91+
92+
### Installation
93+
94+
All examples require you to first install the AgentKit SDK [Installation Reference](https://volcengine.github.io/agentkit-sdk-python/content/1.introduction/2.installation.html)
95+
96+
## Development Guide
97+
98+
### Code Structure
99+
100+
Each example follows the standard AgentKit application structure:
101+
102+
```
103+
Example Directory/
104+
├── agent.py # Agent main program
105+
├── requirements.txt # Dependency list
106+
├── config/ # Configuration files
107+
└── README.md # Detailed instructions
108+
```
109+
110+
### Best Practices
111+
112+
1. **Modular Design**: Separate tools, agents, and configurations
113+
2. **Error Handling**: Implement comprehensive exception handling
114+
3. **Logging**: Use structured logging for easier debugging
115+
4. **Configuration Management**: Use environment variables and config files
116+
117+
## Contributing
118+
119+
We welcome community contributions! If you have new examples or improvement suggestions, please:
120+
121+
1. Fork this repository
122+
2. Create a feature branch (`git checkout -b feature/amazing-example`)
123+
3. Commit your changes (`git commit -m 'Add amazing example'`)
124+
4. Push to the branch (`git push origin feature/amazing-example`)
125+
5. Create a Pull Request
126+
127+
## License
128+
This project is licensed under the [Apache 2.0 License](./LICENSE)
129+
130+
## Support & Feedback
131+
132+
- **Documentation**: Check [AgentKit Official Documentation](https://www.volcengine.com/docs/86681/1844823?lang=zh)
133+
- **Issues**: Report problems in GitHub Issues
134+
135+
## Related Resources
136+
137+
- [AgentKit Official Website](https://www.volcengine.com/docs/86681/1844823?lang=zh)
138+
- [AgentKit SDK/CLI Documentation](https://volcengine.github.io/agentkit-sdk-python/)
139+
- [veadk Official Documentation](https://volcengine.github.io/veadk-python/)
140+
141+
---
142+
143+
**Start exploring the powerful capabilities of AgentKit! Choose an example that interests you, follow the tutorials, and build your own agent applications.**

0 commit comments

Comments
 (0)