aaif-goose / aaif-goose/goose

MCP server worked with 1.49 and don't work with 1.50

未关闭
#11,985 2 条评论 1 个 reaction 已指派 1 人 已被 @jbg 认领 在 GitHub 查看
主要语言
Rust
星标
54.2k
派生
6.2k
平均合并
3 天 4 小时
30 天内合并 PR
240

描述

**Describe the bug**

💡 Before filing, please check common issues:
https://goose-docs.ai/docs/troubleshooting

📦 To help us debug faster, attach your **diagnostics zip** if possible.
👉 How to capture it: https://goose-docs.ai/docs/troubleshooting/diagnostics-and-reporting/

A clear and concise description of what the bug is.

---

I made my own MCP server. (Python3, mcp>=1.0,<2)
It works very good with Goose 1.49 but after I updated to Goose 1.50 it breaks.

Goose shows me an error:
```
failed to initialize MCP client: uncorrelated error response: expected id 0, error response carried server-error
```

My MCP server shows an error:
```
[09/10/26 13:35:39] INFO Created new transport with session ID: a25c5d78f3c84167a1a86e0c86bf663a streamable_http_manager.py:302
INFO: 10.0.0.207:54932 - "POST /macrotracker/mcp HTTP/1.1" 400 Bad Request
```
If you want to see it, it available on https://github.com/DrMoriarty/ru-finance-mcp

---

**Expected behavior**
It worked with 1.49 and I expect that it will works with 1.50 also.

---

**Screenshots**

Image

---

**Please provide the following information**
- **OS & Arch:** MacOS 14.7.4 (23H420)
- **Interface:** UI
- **Version:** 1.50 (latest)
- **Extensions enabled:** My own MCP server
- **Provider & Model:** It doesn't matter

---

**Additional context**
Add any other context about the problem here.

Do not begin implementation until the issue reaches **Ready** on the [Goose Issues board](https://github.com/orgs/aaif-goose/projects/1).

贡献指南

打开贡献指南

调研方向

这个 issue 涉及 Goose 1.49 和 1.50 版本之间的 MCP (Model Context Protocol) 服务器兼容性。首先检查 Goose repository 中的 MCP 客户端初始化代码,可能位于类似 'mcp/' 或 'src/mcp/' 的目录中。查找版本之间错误处理或 ID 关联方面的变更。用户的 MCP 服务器使用 Python 编写,可在 https://github.com/DrMoriarty/ru-finance-mcp 获取;在本地测试交互。 “完成”表示 MCP 服务器能够在 Goose 1.50 中运行,且不会出现 'uncorrelated error response'。

由索引模型根据 Issue 内容生成。

评估

技术栈
python
领域
ai-infra-agents, backend
Issue 类型
缺陷
难度
3/5
预计耗时
1-2 天
活跃度
活跃
描述清晰度
基本清楚
新手友好度
50/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。