neuwirth
/
epiphany
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

CLAUDE.md: drop a live count and disambiguate the citation-gate rule 

Two corrections to the rule added in df5da35.

"This has now happened four times" was a live count in a standing rule, and it
would go stale at the next occurrence — the same defect the rule sits beside, and
the same one this repo keeps paying for in restated test counts. Now count-free:
"This has recurred across multiple contracts."

"Nothing is wrong with the tree" was ambiguous in the one way that matters here,
because two baselines are in play. Nothing tracked has regressed: no committed
file is wrong and HEAD is still green. What is red is the baseline you can
actually observe, because the suite reads the working directory and the untracked
draft is in it. The distinction is the whole point of the rule, so it now says
both halves and adds why it matters anyway: that observable baseline is the one
the mutation discipline depends on, so the effect is the same as a real
regression.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ps1szk2mSfgp4Cz21eVH9x
This commit is contained in:
Levi Neuwirth 2026-08-12 17:00:30 +02:00
parent df5da3512c
commit 552aacac2c
No known key found for this signature in database
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
CLAUDE.md
View File

@ -45,8 +45,11 @@ version literals in `.tex` prose.
is written.** `requirement_labels.rs` scans `spec/` **as files, not as git
objects**, so an *untracked draft* is already in scope: the instant a draft cites
a label the `.tex` does not define, `every_requirement_citation_is_defined` goes
red. Nothing is wrong with the tree, and the baseline the mutation discipline
depends on is gone. This has now happened four times.
red. **Nothing tracked has regressed** — no committed file is wrong, and the
tracked tree at `HEAD` is still green — **but the baseline you can actually
observe is red**, because the suite reads the working directory and the draft is
in it. That is the baseline the mutation discipline depends on, so the effect is
the same as a real regression. This has recurred across multiple contracts.
The standing procedure: