HN Explorer

Ask HN: How to Handle Claude's Stubbornness?

9 points by classicpsy ↗ HN
I’ve been running into this issue with Claude (in browser and claude code) where it is being stubborn many times.

It will refrain from obeying what I say or suggest. Instead, it comes up with its own analogy and starts suggesting and implementing that logic. Even when I already know its going in wrong path and I have proposed the direction for the same problem.

Many times I’ve had to explicitly tell it: don’t think, don’t force your analogy, just implement what I am putting in front of you. Still this does not stick with it always.

Has anyone else dealt with this? How are you handling it?

7 comments

[ 2.7 ms ] story [ 31.4 ms ] thread
ankarhem
Haven't really experienced it besides in long contexts. Usually keep it concise. Something like:

  We're building feature X.
  - You might need `a`, `b`, `c`. (any libraries, url documentation etc)
  - The requirements are:
    - x
    - y
    - z
Any negatives (prevent going down a path) would go into requirement.

This would be a prompt to sisyphus if small or prometheus if big. (Using opencode + oh-my-openagent).

If I believe the agent won't understand what it is supposed to do or if there are multiple solutions of which only some are allowed I add "DO NOT make or edit any business rules before asking me".

LatencyKills
I use a series of stop hook [0] scripts. For example, I have a script [1] that forces Claude to execute tests whenever code files are changed. The stop hook runs automatically and will force Claude to continue working until the script passes.

I also have a script that forces Claude to generate a summary of work [2] if it hadn't done so on it's own.

[0]: https://code.claude.com/docs/en/hooks

[1]: https://gist.github.com/Looking4OffSwitch/c3d5848935fec5ac3b...

[2]: https://gist.github.com/Looking4OffSwitch/3b13b65e40284be899...