Repository navigation
Conversation
Follow-up to vllm-project#62601 review: treat the rules as defaults that can be deviated from with a stated reason, spell out the trade-off between end-to-end and unit tests, allow mocks for objects that are impractical to construct, allow seeded statistical assertions for sampling tests, and prefer adding a case to an existing parametrized test. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: aoshen02 <281812371+aoshen02@users.noreply.github.com>
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?s=60&v=4){kind=small}
6 tasks done
tomasruizt
approved these changes
Oct 22, 2026
tomasruizt
left a comment
tomasruizt
left a comment
Contributor
There was a problem hiding this comment.
thanks for including these!
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Purpose
Follow-up to #62601, addressing @tomasruizt's review (#62601 (review)).
tests/AGENTS.mdare defaults: an agent may deviate when one does not fit, and must say why in the PR. Weakening a test to make it pass stays forbidden.AGENTS.md, prefer adding a case to an existing parametrized test, then extending existing tests, before writing a new one.Test Plan
Docs-only change. Ran
markdownlint-cli2with the repo config on both files. Gave fresh agents with no prior context (a Claude subagent and Kimi) four tasks against this checkout: a top-p edge-case fix, a min-p distribution change at temperature 0.8, a helper inside the KV cache manager, and a 1-in-5 CI flake with anatolmismatch.Test Result
markdownlint-cli2: 0 issues. Both agents added a case to the existing sampler test file, used a seeded chi-square test for the sampling change, tested the KV cache helper as a pure function or through the existing real-manager fixtures instead of mocks, and refused to widenatol, add reruns, orxfailfor the flake.AI assistance (Claude Code) was used to draft this change; the submitter reviewed every changed line.
🤖 Generated with Claude Code