API Service Hangs When Using Agent Node in Chatflow

### Self Checks

- [x] I have read the [Contributing Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) and [Language Policy](https://github.com/langgenius/dify/issues/1542).
- [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones.
- [x] I confirm that I am using English to submit this report, otherwise it will be closed.
- [x] 【中文用户 & Non English User】请使用英语提交，否则会被关闭 ：）
- [x] Please do not modify this template :) and fill in all the required fields.

### Dify version

v1.15.0

### Cloud or Self Hosted

Self Hosted (Docker)

### Steps to reproduce

When I use an Agent node in a chatflow, the entire API service hangs. After the timeout, the API immediately returns to normal. Alternatively, if I restart the plugin daemon, the API reports an error and then returns to normal, as shown in the screenshot below.
1. Use an Agent node.

2. All API services hang.

3. After running for a while, an error occurs.

Error:
Failed to transform agent message: req_id: 429eb69aa3 PluginInvokeError: {"args":{},"error_type":"Exception","message":"invocation exited without response"}

The plugin being used is Agent Strategies.

### ✔️ Expected Behavior

Can call MCP normally without hanging

### ❌ Actual Behavior

MCP is being called (logs are visible from plugin_daemon), but all services hang. The service becomes completely unresponsive with no output, and then eventually returns an error.

Reproduce the hang with an Agent node in a self-hosted Docker deployment, then inspect the plugin_daemon logs around the MCP call and the “invocation exited without response” error. Done means the API remains responsive while the Agent Strategies plugin calls MCP and the request completes or reports a timely error.

Written by the indexing model from the issue text.