Repository navigation
CRAFT v0.4.1: re-pin presmaker + paper-writer to v1.3.1 (Cycle-3 + follow-up) - #1
Merged
Merged
Conversation
…llow-up)
Cycle-3 follow-up coordinated release, Phase 2. Re-pins the two
producer-skill submodules to v1.3.1 and bumps CRAFT 0.4.0 → 0.4.1.
Patch bump — bug fixes in the pinned skills; no contract/schema change
(run-record.v1 is unchanged).
Submodule re-pins:
skills/beril-presentation-maker-skill v1.2.0 → v1.3.1 (4f5b90c)
skills/beril-paper-writer-skill v1.2.0 → v1.3.1 (a6449a9)
skills/beril-adversarial-skill v0.7.1 (unchanged — this
follow-up does not touch adversarial; it is single-shot, no halt,
so neither P0 applies)
Note: the pins jump v1.2.0 → v1.3.1 directly. The CRAFT submodules
were never re-pinned to the v1.3.0 Cycle-3 run-record release, so this
single re-pin brings CRAFT current with BOTH the Cycle-3 run-record.v1
emitter work (Steps 2/4) AND the follow-up boundary fixes:
- P0-1 (presmaker): one run-N allocator — the legacy summary.json
archive (a second uncoordinated _next_run_n) retired.
- P0-2 (presmaker + paper-writer): resume RE-OPENS the canonical
record by status instead of allocating a fresh run-N, so a deck/
manuscript paused at an expected halt stays ONE run record across
the resume (no fragmentation, no craft-status reset to $0).
Family E (cross-skill run-record.v1 conformance): re-pinning brings the
v1.3.1 goldens into the submodule trees, so the presentation-maker and
paper-writer conformance skips flip GREEN. adversarial's Family-E skip
remains (it ships no run-record goldens at v0.7.1 — its Cycle-3 emitter
is a separate, not-yet-released track).
Verify (CC-local):
- craft pytest: 151 passed, 1 skipped (adversarial Family-E golden
only — was 3 skipped pre-re-pin; presmaker + paper-writer now
green). conformance suite: 78 passed, 1 skipped.
- Ruff: 4 PRE-EXISTING findings in cli.py (the `craft status` code
from Step 1 — F841/F541); zero introduced by this re-pin (only
submodule pins + version strings changed).
- Version bumped in BOTH sources: pyproject.toml + src/craft/__init__.py
(0.4.0 → 0.4.1).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.8 (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
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.
Cycle-3 follow-up coordinated release, Phase 2. Re-pins the two producer-skill submodules to v1.3.1 and bumps CRAFT 0.4.0 → 0.4.1 (patch; pinned-skill bug fixes, no contract/schema change).
Re-pins:
beril-presentation-maker-skillv1.2.0 → v1.3.1 (4f5b90c)beril-paper-writer-skillv1.2.0 → v1.3.1 (a6449a9)beril-adversarial-skillv0.7.1 (unchanged — single-shot, neither P0 applies)The pins jump v1.2.0 → v1.3.1 directly (the submodules were never re-pinned to v1.3.0), so this brings CRAFT current with both the Cycle-3 run-record.v1 emitter work AND the follow-up boundary fixes (P0-1 one-allocator, P0-2 resume-re-opens).
Family E: re-pinning brings the v1.3.1 run-record goldens into the submodule trees → presentation-maker + paper-writer conformance skips flip GREEN (adversarial's remains; it ships no goldens at v0.7.1).
Verify: craft pytest 151 passed / 1 skipped (was 3 skipped pre-re-pin — 2 flipped green); zero new ruff (4 pre-existing in cli.py); version bumped in pyproject + init.
🤖 Generated with Claude Code