improvise

Files

T

Edward Langley afbe7e7d1f docs: record the measured generator coverage of the new kind_line rule

context/code/formula-and-io.md already says a new grammar production needs
generator coverage or the proptests quietly stop describing the format.
`kind_line` is an alternative of `category_entry`, so the grammar walk
reaches it with no generator change — but that is precisely the claim the
section exists to say must be checked rather than assumed.

Probed over 500 files from generator::improv_file(): 14 carried a
`kind: label` line and all 14 parsed to a CategoryKind::Label, so each
256-case proptest run exercises the production about seven times.

The probe is not kept. Its rate depends on proptest's internal RNG, so as
a standing assertion it would fail on a dependency bump for a reason that
has nothing to do with the format — the measurement belongs in the
argument, where it can be re-run and corrected.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01FC2jVn8Au7E9JPkw6bwXXH

2026-09-22 01:15:47 -07:00

clients.md

review: the two-facts claim was stated in four places, and I changed one

2026-09-22 00:22:12 -07:00

command.md

fix(command): two P1s in the key path, and both were about silence

2026-09-22 00:13:45 -07:00

core.md

fix: seat a category by its kind, not by its name

2026-09-22 00:57:48 -07:00

doc-site.md

doc-site: name context/code/pr-review.md as an omission