All runs are failing after new release (last night) v0.14.0 #423
Replies: 4 comments 8 replies
-
Hey so sorry thats failing. The MCP errors shouldnt cause any actual issue. Error wasn't the correct naming, I'll sort that. Regards the APICallError. The model is probably trying to read a file which is too long. I have a todo to make a context window manager which should help mitigate some of these issues. Can you run the action with the --debug flag to see the tools called? I would advise you to pin the action version to a published tag and also to upgrade to gpt-4.1-mini as it's cheaper than 4o and better at instruction following. |
Beta Was this translation helpful? Give feedback.
-
Thanks @mattzcarey ok, so using gpt-4.1-mini helped with -
here's the debug snippet
|
Beta Was this translation helpful? Give feedback.
-
for now I will continue to use |
Beta Was this translation helpful? Give feedback.
-
Hey, v0.14.2 release fixes the github action.yml parsing and adds it as a test to CI to stop this happening again. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
I have an action job configured like this;
I even have this configured to use v0.10.0, but it seems that shippie is pulling latest - that's a different issue, but it kinda leads into this breakage.
the config above has not changed. all PRs were being processed without error - until now.
here's the log snippet
this would imply that the MCP config is required?
Beta Was this translation helpful? Give feedback.
All reactions