super-ux 0.62.0 → 0.63.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +22 -0
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,3 +1,25 @@
|
|
|
1
|
+
## 0.63.0 (2026-10-20): short descriptions, triggers in the body
|
|
2
|
+
|
|
3
|
+
- **Every skill description is a router of 279 to 393 characters** (was 332 to
|
|
4
|
+
948). The operator's Claude Code skill listing measured 29,995 of about
|
|
5
|
+
30,000 characters for 134 skills, and 81 skills showed no description at
|
|
6
|
+
all; a plugin skill cannot be hidden, so length is the only lever. Each
|
|
7
|
+
description is one "Use when" sentence, four to six English and Russian
|
|
8
|
+
triggers, one "Not for" boundary naming the sibling, and the refusal
|
|
9
|
+
phrase where the route has one. Total for the seven skills: 4,426 to 2,436.
|
|
10
|
+
- **Every removed trigger phrase moved verbatim into a `## Triggers` section**
|
|
11
|
+
of its SKILL.md body, so the family routing hook's words still appear in the
|
|
12
|
+
skill's own text.
|
|
13
|
+
- **`ux-flows` Improve procedure moved to `references/flow-improve.md`**, text
|
|
14
|
+
unchanged apart from link paths, so the body stays inside the house audit's
|
|
15
|
+
4,750-token working limit with the new `## Triggers` section (4,712 tokens
|
|
16
|
+
before, 4,564 after).
|
|
17
|
+
- **Commands carry one short sentence each** (60 to 89 characters, was 99 to
|
|
18
|
+
224; total 2,424 to 1,116).
|
|
19
|
+
- **`validate.py` holds the budget:** a skill description over 500 characters
|
|
20
|
+
or without a "Not for" boundary fails, a command description over 120 fails,
|
|
21
|
+
and the trigger-position check reads `Triggers:` as well as `Triggers - `.
|
|
22
|
+
|
|
1
23
|
## 0.62.0 (2026-10-19): retrieval crawlers, the audit's brief, landing promises, Copylot sweep
|
|
2
24
|
|
|
3
25
|
- **`B050` judges retrieval agents only.** The AI-search contradiction is a block
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "super-ux",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.63.0",
|
|
4
4
|
"scripts": {
|
|
5
5
|
"test": "python3 test/validate.py && python3 test/brand_lint_test.py && python3 test/display_copy_test.py && python3 test/ux_lint_test.py && python3 docs/ux/lint.py && python3 docs/brand/lint.py && node test/installer_test.js && npm run test:audit",
|
|
6
6
|
"test:audit": "for t in test/audit_regressions/*.py; do python3 \"$t\" || exit 1; done"
|