v1.1.0 release prep#5
Merged
Merged
Conversation
Pin the GitHub Action's pip install to skillcheck>=1.0.1 so consumers fail loudly on unpublished v1 features instead of silently running v0.2.0. Replace seven gitignored runs/* path citations in the README with the exact commands needed to reproduce each finding. Delete the unrelated git-commit-crafter SKILL.md from the repo root (canonical example lives at skills/skillcheck/SKILL.md). Drop the false @v0 tag claim from the README (only @v0.2.0 was ever pushed) and update the CHANGELOG to point at the action's pinned tag. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Drop comprehensive, robust, and flexible from _VAGUE_WORDS. Each can
describe a concrete attribute when qualified ("comprehensive coverage of
N formats", "robust against malformed input"), so flagging them
penalizes legitimate descriptions. Document the inclusion rubric in a
comment so future additions stay grounded.
Collapse _ACTION_VERBS from 86 entries (base + 3rd-person duplicates)
to 42 base forms. Add _is_action_verb() to handle stem normalization
across -s, -es, and -ies endings, so adding a new verb only requires
the base form. Two regression tests guard the new rubric: comprehensive
alone no longer drops the score, and 3rd-person verbs still register
without explicit enumeration.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Exit code 2 was conflating tool-misuse errors (missing path, conflicting flags) with valid runs that produced only warnings, leaving CI consumers unable to tell them apart. Warning-only runs now exit 0 by default. The new --warnings-as-errors flag escalates warnings to exit 1 for stricter gates. Code 2 is now reserved for input errors only. The README exit-code table and the affected tests are updated in the same commit. main() previously ran 428 lines, far past the project's module-size discipline. Extract the conflict checks, show-history dispatch, emit-mode dispatch, validation pipeline, exit-code computation, history recording, and report printing into named helpers (_check_flag_conflicts, _handle_show_history, _dispatch_emit_mode, _run_validation, _compute_exit_code, _record_history, _print_report, plus _setup_io_encoding and _validate_arg_choices). main() is now 52 lines of linear top-level dispatch. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The audit's claim 7 stretch goal was main() under 100 lines and cli.py under 700. Extracting nine helpers got main() to 52 lines but pushed cli.py from 1127 to 1172 because each helper added a signature plus a docstring without a corresponding inline-code reduction. The audit explicitly allowed treating the file size as a deliberate choice rather than refactoring; the helper extraction made the total-lines metric worse, so revert it. Restores main() to its pre-audit inline form. Keeps the --warnings-as-errors flag and the exit-code split from claim 4 (warning-only -> 0 by default, 1 with --warnings-as-errors, code 2 reserved for input errors). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The v1.0.1 -> v1.1.0 jump captures: (a) the audit fixes from this session (action.yml pin, README runs/* citations, top-level SKILL.md deletion, false @v0 claim, description scorer rubric); (b) the exit-code split which is a behavior reversal of v1.0.1 (warning-only is now 0, not 2); (c) the new --warnings-as-errors flag. The first set is patch- level housekeeping but the exit-code change and the new flag are public-surface changes; semver demands a minor bump. CHANGELOG entry consolidates the prior [Unreleased] section into 1.1.0. RELEASE_NOTES_v1.1.0.md documents the migration path for any CI that was depending on v1.0.1's "warnings exit 2" behavior. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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
1 participant
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.
Summary
cli.pyleft at pre-audit size after the helper-extraction attempt grew the file overall)--warnings-as-errorsflag escalates warning-only runs to 1 for stricter CI gatesanthropics/skillsshows zero score deltas across 17 filesCHANGELOG.md1.1.0 section andRELEASE_NOTES_v1.1.0.mdadded; rubric-change wording tightened to reflect the no-op-against-corpus finding instead of implying false positives in the wildTest plan
RELEASE_NOTES_v1.1.0.mdand CHANGELOG wording matches the corpus-verification finding (no-op against current corpus; new tests are forward-looking guards)