open/repo-standard
1
0
Fork 0
Code Issues Releases 2 Activity

docs: test count 32 -> 34, and record the H1 softening

Browse source 
The H1 change added two tests; the badge, the README line and the
CHANGELOG still said 32. Exactly the drift this repo exists to catch,
in this repo - and check-versions.mjs would not have caught it, since
it only reads the version badge.

Also states in the README what actually fails the first-screen check
now that a differing H1 is a WARN.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WYJ3FHLtVgzFXMZ6UF598h
This commit is contained in:
Kjell Tore Guttormsen 2026-07-27 09:19:43 +02:00
commit 9a72cefefa
2 changed files with 8 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

6
CHANGELOG.md
View file

@ -25,7 +25,7 @@ First release. Covers the checks that a single repository can answer on its own.
- **The skill** (`skills/repo-standard/`) — the judgement the script cannot
encode: what a description should say, why the summary card must never be used
to verify one, what not to retrofit, and where the per-repo boundary is.
- 32 tests over the pure classifiers, using the measured false positives as
- 34 tests over the pure classifiers, using the measured false positives as
reference fixtures.
### Notes
@ -38,6 +38,10 @@ First release. Covers the checks that a single repository can answer on its own.
form as the defect.
- The org listing is read in **one** call, anonymously. Per-repo fetching trips
the forge's rate limiter.
- A README H1 that differs from the repo name is a `WARN`, not an `ERROR`. The
thread that has to hold is description == catalog == opening line, and the H1
is none of those three; a human title is a naming choice the operator owns.
A missing H1 is still an `ERROR`.
- No hook ships in this release. A blocking gate has to be precise enough not to
fail a correct repository first.

6
README.md
View file

@ -13,7 +13,7 @@ checks that surface in one repository and reports what it finds.
![Version](https://img.shields.io/badge/version-0.1.0-blue)
![Platform](https://img.shields.io/badge/platform-Claude_Code_Plugin-purple)
![Skills](https://img.shields.io/badge/skills-1-orange)
![Tests](https://img.shields.io/badge/tests-32-green)
![Tests](https://img.shields.io/badge/tests-34-green)
![License](https://img.shields.io/badge/license-MIT-lightgrey)
## Install
@ -52,7 +52,7 @@ It checks five things, and the repository's **class** decides what each one mean
| Check | What fails it |
| --- | --- |
| First screen | `# <name>` is not line 1, or the line under it is not the published description |
| First screen | no H1 on line 1, or the line under it is not the published description. An H1 that merely differs from the repo name is a `WARN` — that is a naming choice, not a defect |
| Install block | the form for this class is missing, incomplete, or shown over `ssh://` |
| Required files | a file this class needs is absent |
| References | an `open/<name>` in URL position resolves to nothing |
@ -112,7 +112,7 @@ without that, a raw scan turns three dead names into about twenty.
npm test
```
32 tests over the pure classifiers. The reference fixtures are the measured
34 tests over the pure classifiers. The reference fixtures are the measured
false positives that produced the three-outcome rule, each with its expected
verdict.