skillshelf 0.7.0 → 0.7.1
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/README.md +3 -6
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,11 +1,8 @@
|
|
|
1
1
|
# skillshelf
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
[](./LICENSE) [](https://bun.sh) [](https://github.com/Wang-Cankun/skillshelf/actions) [](https://www.npmjs.com/package/skillshelf)
|
|
4
4
|
|
|
5
|
-
|
|
6
|
-
[](https://bun.sh)
|
|
7
|
-
[](https://github.com/Wang-Cankun/skillshelf/actions)
|
|
8
|
-
[](https://www.npmjs.com/package/skillshelf)
|
|
5
|
+
**A package manager for your agent skills — one canonical library, loaded on demand, never all at once.**
|
|
9
6
|
|
|
10
7
|
Your skills are scattered across **every agent you use** — `~/.claude/skills`, `~/.codex/skills`,
|
|
11
8
|
`~/.cursor/skills`, Obsidian vaults, and a dozen per-project `.claude` dirs. Copies drift, you
|
|
@@ -222,7 +219,7 @@ Domains and tags can be inferred and re-run by an LLM via `skl infer`. **This is
|
|
|
222
219
|
the entire core (search, ls, show, use, bundles, add, update) works fully without any LLM.**
|
|
223
220
|
Inference has three mutually exclusive modes:
|
|
224
221
|
|
|
225
|
-
```
|
|
222
|
+
```text
|
|
226
223
|
skl infer [--emit | --apply <file.json> | --provider <name>] \
|
|
227
224
|
[--base-url <url>] [--model <id>] [--include-retired] [--json]
|
|
228
225
|
```
|