@victortomaili/skill-cli 0.6.3 → 0.6.4
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 +7 -0
- package/README.md +2 -0
- package/package.json +1 -1
- package/src/lib/agents-md.js +3 -0
package/CHANGELOG.md
CHANGED
|
@@ -11,6 +11,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
|
|
|
11
11
|
- Cursor adapter (`.cursor/rules` format) for `init -g` bootstrap.
|
|
12
12
|
- Per-agent hook adapters for automatic `/X` triggering (push model).
|
|
13
13
|
|
|
14
|
+
## [0.6.4] - 2026-07-08
|
|
15
|
+
|
|
16
|
+
### Added
|
|
17
|
+
- AGENTS.md START GATE: a PRIORITIES line — when trading off how to do the work,
|
|
18
|
+
rank CORRECTNESS first, then QUALITY, then COST (cheap / token-efficient), then
|
|
19
|
+
SPEED (fast); never sacrifice correctness or quality for speed or cost.
|
|
20
|
+
|
|
14
21
|
## [0.6.3] - 2026-07-07
|
|
15
22
|
|
|
16
23
|
### Changed
|
package/README.md
CHANGED
|
@@ -206,6 +206,8 @@ instruction file (`CLAUDE.md`, `AGENTS.md`, `GEMINI.md`):
|
|
|
206
206
|
> *experience* — response **style/format/mode**, **context-cost** optimization, or an
|
|
207
207
|
> optional **shortcut** (easier/faster, not about correctness). When unsure, propose.
|
|
208
208
|
> **LOADED ≠ LISTED**: a skill is loaded only if you `cat`-ed it this session.
|
|
209
|
+
> **PRIORITIES**: correctness > quality > cost (cheap) > speed — never sacrifice
|
|
210
|
+
> correctness/quality for speed or cost.
|
|
209
211
|
> `/X` → `skill trigger X`.
|
|
210
212
|
|
|
211
213
|
It's wrapped in `<!-- BEGIN skill-cli --> … <!-- END skill-cli -->` markers, never
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@victortomaili/skill-cli",
|
|
3
|
-
"version": "0.6.
|
|
3
|
+
"version": "0.6.4",
|
|
4
4
|
"description": "Cross-agent skill manager — one global store, clean agent folders, activation via skill.config",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "Victor Tomaili <victor@tomaili.com>",
|
package/src/lib/agents-md.js
CHANGED
|
@@ -38,6 +38,9 @@ Read it, then for EACH skill decide FROM ITS DESCRIPTION:
|
|
|
38
38
|
• CONTEXT-COST / token optimization;
|
|
39
39
|
• an optional ACCELERATOR / shortcut (easier or faster, but not about correctness).
|
|
40
40
|
When unsure → PROPOSE (proposing is the safe default; loading is silent).
|
|
41
|
+
PRIORITIES — when trading off how to do the work, rank: CORRECTNESS first, then
|
|
42
|
+
QUALITY, then COST (cheap / token-efficient), then SPEED (fast). Never sacrifice
|
|
43
|
+
correctness or quality for speed or cost.
|
|
41
44
|
LOADED ≠ LISTED. A skill is loaded ONLY if you have run \`skill cat <name>\` for
|
|
42
45
|
it THIS session. Seeing a skill in \`skill active\` / \`skill list\` does NOT load
|
|
43
46
|
it — you then know only its name + description, nothing more. Never infer a skill
|