Replies: 1 comment
-
|
I am experiencing this as we speak. Trying to determine at what point should I use spec-kit based on task complexity. I am trying to break things down into smaller changes, inherently less complex, but still should be 'spec'd' out and documented. I want to use the great features of spec-kit but have it be more efficient based on stated or analyzed complexity. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I’d like to propose an enhancement to spec-kit’s planning process: the ability to assess the complexity of a requested feature and adjust the effort level accordingly.
For example, I recently asked spec-kit to add Mermaid and table rendering to a markdown field on my site, spec-kit spent about 45 minutes generating a detailed plan and multiple tasks, but none of it worked and it introduced thousands of new errors. In the end, the fix was just a quick addition of a couple of plugins to my existing markdown renderer (which Claude managed first time with a simple prompt).
It seems that for very simple features, spec-kit can sometimes over-engineer the solution rather than generating a simple task. It would be helpful if spec-kit could:
This would make spec-kit faster, more accurate, and less prone to over-complication for straightforward tasks.
I think it's important that spec-kit can manage a variety of complexity of user features. If it cannot the user will have to make a gut decision if somethint id "big enough" for a spec, or should be handled directly by Claude/Gemini/codex etc.
Beta Was this translation helpful? Give feedback.
All reactions