Repository navigation
chore: delete the dead /implement pathway and three unused scripts - #34
Merged
Merged
Conversation
Upstream retired all of this and aws-samples carries it only because it predates the initial copy: - .claude/skills/implement/ and .claude/skills/summarize/ were deleted upstream in fa63de9, the commit that introduced the security-check skill. - benchmarks/scripts/select-best-lld.sh, extract-metrics.py and replay-benchmark.py were deleted upstream in 4b0bd30, "Remove scripts unused by the vLLM serving and SWE benchmark flows". The three /implement files are one coupled unit, which is why they go together: the skill reads a spec.md that select-best-lld.sh writes, so neither half is dead on its own, but the pathway as a whole has no entry point here. Verified unreferenced before removing: `/summarize` appears nowhere in the tree, and `/implement` only in two passing mentions in the vLLM model guides, which now name /swe2 and /swe3 instead. select-best-lld.sh, extract-metrics.py and replay-benchmark.py are referenced by nothing but each other. Also drops the artifact layout the /implement skill assumed (swe-benchmark-data/<repo>/<problem>/implementations/), which has not matched the harness since artifacts moved to <model>/<harness>/<skill>/<repo>/<task>/. 530 tests pass with no new failures, ruff clean, and a tree-wide grep for every deleted filename now returns nothing.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
All five files were retired upstream and exist here only because they predate the initial copy:
.claude/skills/implement/SKILL.mdfa63de9— the commit that introducedsecurity-check.claude/skills/summarize/SKILL.mdfa63de9benchmarks/scripts/select-best-lld.sh4b0bd30"Remove scripts unused by the vLLM serving and SWE benchmark flows"benchmarks/scripts/extract-metrics.py4b0bd30benchmarks/scripts/replay-benchmark.py4b0bd30The three
/implementfiles are one coupled unit, which is why they go together rather than one at a time: the skill reads aspec.mdthatselect-best-lld.shwrites, so neither half is dead in isolation — but the pathway as a whole has no entry point in this repo.It's also stale in a way that can't be fixed cheaply: the skill assumes an artifact layout (
swe-benchmark-data/<repo>/<problem>/implementations/) that hasn't matched the harness since artifacts moved to<model>/<harness>/<skill>/<repo>/<task>/.Verified unreferenced before removing
/summarizeas a command: zero occurrences tree-wide./implementas a command: only two passing mentions, inself-hosted/vllm/models/glm-5.2.mdandglm-5.3.md, describing which benchmarks cache well. Both now name/swe2and/swe3, which exist.select-best-lld,extract-metrics,replay-benchmark: referenced by nothing but each other.Note the first grep for
implement/summarizelooks alarming — it hits eight skills — but those are the English words, not command references. The slash-prefixed search above is the real one.Checks
530 tests pass, no new failures (the one failure,
test_judge_common.test_lowercase_preferred_on_collision, fails onmaintoo — it assumes a case-sensitive filesystem).ruffclean. A tree-wide grep for every deleted filename now returns nothing.Net −1,077 lines.