Skip to content

Align generated template with conventions; promote universal carry-overs - #11

Merged
dmccoystephenson merged 2 commits into
mainfrom
feature/align-with-conventions
May 23, 2026
Merged

dmccoystephenson merged 2 commits into
mainfrom
feature/align-with-conventions

Conversation

@dmccoystephenson

@dmccoystephenson dmccoystephenson commented May 23, 2026 •

Copy link
Copy Markdown
Member

Summary

This is the first PR in a series bringing all dev-loop skills into line with the conventions just codified in example-skills-catalog/CONVENTIONS.md. Subsequent PRs against example-h-dev-loop, example-k-dev-loop, kingdom-first-era-dev-loop, and example-a-dev-loop will retrofit the same Phase 9 structure into existing generated skills.

Test plan

  • Visual review of create-dev-loop.md Phase 3, Phase 9, Step 4, and Edge cases sections
  • Confirm the new Phase 3 universal rules read clearly without contradicting {{CODE_PATTERNS}} substitution
  • Confirm the new flaky-test entry doesn't conflict with the existing "Tests fail during implementation" entry (they're complementary — that one is for real failures, this one is for re-run-resolvable flakes)

Closes #9
Closes #10


drafted by Claude on behalf of Daniel Stephenson

- Phase 9 generates a numbered structure with duplicate-check on step 1 (#9)
- Step 4 substitution-table preamble notes it covers {{#if FLAG}} conditionals too (#10)
- Phase 3 gains universal "match sibling structure" + "rename siblings together" rules
  (promoted from kingdom-first-era-dev-loop#1 and #2 — they're not KFE-specific)
- Edge cases gain a flaky-test entry (promoted from dpm-dev-loop#12 — every tests-running
  loop will hit this eventually)

Closes #9
Closes #10

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

@dmccoystephenson dmccoystephenson left a comment

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Self-review: found one bug, fixed in 0ff1aea — the new gh issue create block in Phase 9 used \ (double backslash) for line continuations instead of the single \ used everywhere else in the file. Would have rendered as literal \ in the generated skill and broken shell parsing. No other issues found on a second pass over the diff.

@dmccoystephenson
dmccoystephenson merged commit 0df08e6 into main May 23, 2026
@dmccoystephenson
dmccoystephenson deleted the feature/align-with-conventions branch May 23, 2026 23:45
The new gh issue create block in the generated Phase 9 used \\ (double
backslash) for line continuations, which would render as literal \\ in
the generated skill and break the command in shell. Match the convention
used by every other line continuation in this file (single \).
dmccoystephenson added a commit that referenced this pull request Aug 3, 2026
…ers (#11)

* Align generated template with conventions; promote universal carry-overs

- Phase 9 generates a numbered structure with duplicate-check on step 1 (#9)
- Step 4 substitution-table preamble notes it covers {{#if FLAG}} conditionals too (#10)
- Phase 3 gains universal "match sibling structure" + "rename siblings together" rules
  (promoted from kingdom-first-era-dev-loop#1 and #2 — they're not KFE-specific)
- Edge cases gain a flaky-test entry (promoted from a-private-repo-1#12 — every tests-running
  loop will hit this eventually)

Closes #9
Closes #10

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* Self-review fix: use single \ for line continuations in Phase 9 template

The new gh issue create block in the generated Phase 9 used \\ (double
backslash) for line continuations, which would render as literal \\ in
the generated skill and break the command in shell. Match the convention
used by every other line continuation in this file (single \).

---------

Co-authored-by: BuildTools <unconfigured@null.spigotmc.org>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

1 participant