[Bug]: Speckit skills are not interactive
- 主要语言
- Python
- 星标
- 137k
- 派生
- 12.3k
- 平均合并
- 2 天 12 小时
- 30 天内合并 PR
- 159
描述
### Bug Description
I am using spec-kit latest version 1.0.6 for 3 projects. These are the problems I faced.
1. Most of the skills don't ask questions from the user. It just start doing it's join as and when I submit the /speckit-*** command
2. In my recent case there were 60 tasks that were documented by /speckit-tasks. But when I tried /speckit-implement command it' straight away implemented 25 tasks until I realized that it's continuously implementing the tasks without asking me for review. My point is the command should be interactive like implement task by task and then ask user's review and user's confirmation before proceeding to implement next task.
### Steps to Reproduce
I tried these commands in the order FYI
1. /speckit-constitution
2. /speckit-specify
3. /speckit-plan
4. /speckit-tasks
5. /speckit-implement
### Expected Behavior
1. /speckit-*** skills should be interactive with the user and ask questions to user to get more context
2. /speckit-*** skills should implement task by task and then ask user's review and user's confirmation before proceeding to implement next task.
### Actual Behavior
1. /speckit-*** skills are behaving mostly autonomous and not asking user's inputs for getting more context.
2. /speckit-*** skills are not asking for user's review and continually implementing on it's own.
### Specify CLI Version
1.0.6
### AI Agent
GitHub Copilot
### Operating System
Windows 11
### Python Version
Python 3.13.15
### Error Logs
```shell
No error logs
```
### Additional Context
_No response_
贡献指南
调研方向
Reproduce the reported flow with /speckit-constitution, /speckit-specify, /speckit-plan, /speckit-tasks, and /speckit-implement using spec-kit 1.0.6 and GitHub Copilot. Review how these command entry points handle user prompts and task progression; done means skills request needed context and /speckit-implement pauses for review and confirmation between tasks.
由索引模型根据 Issue 内容生成。
评估
- 技术栈
- python
- 领域
- ai, cli, tooling
- Issue 类型
- 缺陷
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 活跃度
- 活跃
- 描述清晰度
- 基本清楚
- 新手友好度
- 35/100