a2a-agent-mcpserver-generator 0.1.1__tar.gz → 0.1.2__tar.gz

{a2a_agent_mcpserver_generator-0.1.1 → a2a_agent_mcpserver_generator-0.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: a2a-agent-mcpserver-generator
-Version: 0.1.1
+Version: 0.1.2
 Summary: Add your description here
 Author-email: vincent-pli <justdoit.pli@gmail.com>
 Requires-Python: >=3.13
@@ -24,6 +24,11 @@ English | [简体中文](./README-zh.md)
 - Complete Project Setup: Generates all necessary files to run an MCP server
 - Easy Configuration: Simple environment-based configuration for the generated server
 
+## Prerequisites
+
+- Python 3.13+
+- `uv` (optional, but recommended) or `pip`
+
 ## Environment Setup 🔧
 Firstly, you need to start up a a2a agent:
 https://github.com/google/a2a-python/tree/main/examples
@@ -35,7 +40,7 @@ uv run a2a-agent-mcpserver-generator --agent http://0.0.0.0:10000
 ## Installation 📦
 
 ```bash
-pip install a2a-agent-mcpserver-generator
+uv pip install a2a-agent-mcpserver-generator
 ```
 
 ## Usage 🚀

a2a_agent_mcpserver_generator-0.1.2/README-zh.md

@@ -0,0 +1,61 @@
+# A2A Agent 到 MCP 服务器 🤖
+这是一个命令行工具，可以从一行命令的 a2a 代理生成 Model Context Protocol（MCP）服务器代码。这个工具可以帮助你快速创建一个 MCP 服务器，用来作为 LLMs（大语言模型）和 a2a 代理之间的桥梁。
+English | [简体中文](./README-zh.md)
+
+## 特性 ✨
+- 自动工具生成：将 a2a 代理的每个 `技能` 转换为一个 MCP 工具
+- 传输选项：仅支持 stdio，如果要使用 sse 可以借助 mcp-proxy
+- 完整的项目设置：生成运行 MCP 服务器所需的所有文件
+- 简易配置：生成的服务器采用基于环境的简单配置
+
+## 前置条件
+- Python 3.13+
+- `uv` (可选，但建议) 或 `pip`
+
+## 环境设置 🔧
+首先，你需要启动一个 a2a 代理：
+https://github.com/google/a2a-python/tree/main/examples
+```
+uv run a2a-agent-mcpserver-generator --agent http://0.0.0.0:10000
+```
+## 安装 📦
+```bash
+uv pip install a2a-agent-mcpserver-generator
+```
+## 使用 🚀
+```
+uv run a2a-agent-mcpserver-generator --help
+使用方法：a2a-agent-mcpserver-generator [选项]
+选项:
+  --agent TEXT
+  --output TEXT
+  --name TEXT
+  --history BOOLEAN
+  --use_push_notifications BOOLEAN
+  --push_notification_receiver TEXT
+  --help                          显示此消息并退出。
+```
+## E2E 示例
+建议使用 [mcpclihost](https://github.com/vincent-pli/mcp-cli-host) 作为 MCP 主机进行尝试。
+这个工具(`mcpclihost`)可以同时支持 Azure Openai 和 deepseek
+你可以像这样添加生成的 MCP 服务器配置：
+```
+{
+  "mcpServers": {
+    "a2a-mcp": {
+      "command": "uv",
+      "args": [
+        "--project",
+        "/Users/lipeng/workspaces/github.com/vincent-pli/a2a-agent-mcpserver-generator/a2a-mcp-server",
+        "run",
+        "a2a-agent-mcpserver"
+      ]
+    }
+  }
+}
+```
+向 `~/.mcp.json`（`mcpclihost` 的默认 MCP 服务器配置路径）添加上述内容，然后试试看。
+
+## 许可协议
+
+这个项目采用 Apache 2.0 许可协议 - 详情请见[LICENSE](LICENSE)文件。

{a2a_agent_mcpserver_generator-0.1.1 → a2a_agent_mcpserver_generator-0.1.2}/README.md

@@ -11,6 +11,11 @@ English | [简体中文](./README-zh.md)
 - Complete Project Setup: Generates all necessary files to run an MCP server
 - Easy Configuration: Simple environment-based configuration for the generated server
 
+## Prerequisites
+
+- Python 3.13+
+- `uv` (optional, but recommended) or `pip`
+
 ## Environment Setup 🔧
 Firstly, you need to start up a a2a agent:
 https://github.com/google/a2a-python/tree/main/examples
@@ -22,7 +27,7 @@ uv run a2a-agent-mcpserver-generator --agent http://0.0.0.0:10000
 ## Installation 📦
 
 ```bash
-pip install a2a-agent-mcpserver-generator
+uv pip install a2a-agent-mcpserver-generator
 ```
 
 ## Usage 🚀

{a2a_agent_mcpserver_generator-0.1.1 → a2a_agent_mcpserver_generator-0.1.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "a2a-agent-mcpserver-generator"
-version = "0.1.1"
+version = "0.1.2"
 description = "Add your description here"
 readme = "README.md"
 authors = [

{a2a_agent_mcpserver_generator-0.1.1 → a2a_agent_mcpserver_generator-0.1.2}/src/a2a_agent_mcpserver_generator/app.py

@@ -45,9 +45,14 @@ async def main(
     log.info(f'- Server Name: {name}')
 
     # Load agent.json
-    async with httpx.AsyncClient() as httpx_client:
-        card_resolver = A2ACardResolver(httpx_client=httpx_client, base_url=agent)
-        card: AgentCard = await card_resolver.get_agent_card()
+    try:
+        async with httpx.AsyncClient() as httpx_client:
+            card_resolver = A2ACardResolver(httpx_client=httpx_client, base_url=agent)
+            card: AgentCard = await card_resolver.get_agent_card()
+    except Exception as e:
+        log.error(f"failed to get agent card from address: {agent}")
+        log.error(e)
+        return
 
     # Parse agent.json
     card_parsed: CardParsed = parse_card(card)

{a2a_agent_mcpserver_generator-0.1.1 → a2a_agent_mcpserver_generator-0.1.2}/src/a2a_agent_mcpserver_generator/server_generator.py

@@ -145,9 +145,10 @@ async def handle_call_tool(name: str, arguments: dict) -> list[types.TextContent
 
                 if isinstance(chunk.root.result, TaskStatusUpdateEvent):
                     task_status = chunk.root.result
-                    ctx.session.send_log_message(
+                    notification_msg = ''.join(part.root.text for part in task_status.status.message.parts) if task_status.status.message else ""
+                    await ctx.session.send_log_message(
                         level="info",
-                        data=f"Task: {{task_id}} is {{task_status.status.state}} at {{task_status.status.timestamp}} with message: {{task_status.status.message}}",
+                        data=f"Task: {{task_id}} is '{{task_status.status.state}}' with message: {{notification_msg}}",
                         logger="notification_stream",
                         related_request_id=ctx.request_id,
                     )
@@ -188,10 +189,10 @@ async def handle_call_tool(name: str, arguments: dict) -> list[types.TextContent
                             )]
                         
                         task = get_response.root.result
-
-                        ctx.session.send_log_message(
+                        notification_msg = ''.join(part.root.text for part in task_status.status.message.parts) if task_status.status.message else ""
+                        await ctx.session.send_log_message(
                             level="info",
-                            data=f"Task: {{task_id}} is {{task.status.state}} at {{task.status.timestamp}} with message: {{task.status.message}}",
+                            data=f"Task: {{task_id}} is '{{task_status.status.state}}' with message: {{notification_msg}}",
                             logger="notification_stream",
                             related_request_id=ctx.request_id,
                         )