Improve benchmark task error messages in kaggle CLI#1057
Merged
Conversation
The status command was hiding `creation_error_message`, so users had no way to see why a task failed. Surface it inline below the status line. Also replace the raw `.title()` rendering of enum names (which produced awkward output like `Kernel_Without_Run`) with a small mapping to human-friendly labels, falling back to titlecased words for unknown states. Co-authored-by: kaggle-agent <kaggle-agent@users.noreply.github.com>
The previous failure labels were terse ("Failed — No run output") and
didn't tell users how to fix the problem. Expand each label to name the
likely cause and point at the next step (e.g. calling .run(), passing a
model via kbench.llm, checking the notebook log), so users can act on
the status output without digging through docs.
Co-authored-by: kaggle-agent <kaggle-agent@users.noreply.github.com>
dolaameng
reviewed
Jun 11, 2026
The status command now renders friendly creation-state labels and an Error line when the task itself failed, but the user docs and skill reference still showed the old raw-enum example output. Update them so users (and the skill-driven agent) see what the CLI actually prints. Co-authored-by: kaggle-agent <kaggle-agent@users.noreply.github.com>
Baking human-readable explanations into `_STATE_LABELS` mixed two things the CLI should keep distinct: the failure *kind* (the enum) and the failure *message* (the server's `creation_error_message`). It also meant the CLI had to be updated every time the server added a new failure mode. Drop the label map, render the cleaned enum as-is, and rely on the server's `creation_error_message` for the explanation. Apply the same pattern to `kaggle b t run` so its refusal-to-run error follows the same `status: <KIND>` + `Error: <message>` shape as `status` and the push/run wait paths. Co-authored-by: kaggle-agent <kaggle-agent@users.noreply.github.com>
dolaameng
approved these changes
Jun 11, 2026
Contributor
stevemessick
approved these changes
Jun 11, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When a benchmark task failed, the CLI hid the server's
creation_error_messageand rendered the failure as a raw titlecasedenum, so users had no idea what went wrong or how to fix it. Surface the
server message as a separate
Error:line and apply the samestatus: <KIND>+Error: <message>shape consistently acrosskaggle b t status,kaggle b t run, and the push/run wait paths —keeping the CLI a thin display layer over the server's classification
and explanation.
Task: limagoog-20260611165119-dba58d70
Context: https://chat.kaggle.net/kaggle/pl/ki55zezbmfg19b3km7ztp3fsww