agent-skill-manager 1.21.0 → 2.0.0
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/README.md
CHANGED
|
@@ -275,7 +275,7 @@ asm publish --yes ./my-skill
|
|
|
275
275
|
4. Test with your AI agent
|
|
276
276
|
5. **Security audit** — `asm audit security awesome-skill`
|
|
277
277
|
6. **Verify metadata** — `asm inspect awesome-skill`
|
|
278
|
-
7. **Score quality** — `asm eval ./awesome-skill`
|
|
278
|
+
7. **Score quality** — `asm eval ./awesome-skill`
|
|
279
279
|
8. Push to GitHub
|
|
280
280
|
9. **Verify install flow** — `asm install github:you/awesome-skill`
|
|
281
281
|
10. **Publish to registry** — `asm publish ./awesome-skill`
|
|
@@ -318,36 +318,25 @@ asm index search "your-skill" --json
|
|
|
318
318
|
|
|
319
319
|
Each indexed skill in the output JSON includes `"verified": true` or `"verified": false`. If verification fails, the ingestion debug log (set `ASM_DEBUG=1`) prints the specific reasons.
|
|
320
320
|
|
|
321
|
-
###
|
|
321
|
+
### Quality Evaluation (`asm eval`)
|
|
322
322
|
|
|
323
|
-
Static verification tells you the SKILL.md is well-formed. `asm eval` goes further and
|
|
324
|
-
|
|
325
|
-
1. **Is it well-written?** — `quality@1.0.0` ships by default and runs a scored rubric over structure, frontmatter, clarity, and safety.
|
|
326
|
-
2. **Does it actually work?** — `asm eval --runtime` shells out to [skillgrade](https://github.com/mgechev/skillgrade) for deterministic + LLM-judge runtime evals in a Docker sandbox, with CI-ready exit codes.
|
|
327
|
-
|
|
328
|
-
**Zero-setup install:** skillgrade ships as a direct dependency of `asm`. After `npm install -g agent-skill-manager`, `asm eval --runtime` just works — no `npm i -g skillgrade`, no PATH hijacking. Override with `ASM_SKILLGRADE_BIN=/path/to/skillgrade` if you want to point at a different binary.
|
|
323
|
+
Static verification tells you the SKILL.md is well-formed. `asm eval` goes further: the built-in `quality` provider runs a scored rubric over structure, frontmatter, clarity, prompt engineering, context efficiency, safety, testability, and naming — and emits per-category scores plus concrete suggestions for improvement. Zero setup, zero API keys, zero external binaries.
|
|
329
324
|
|
|
330
325
|
```bash
|
|
331
|
-
#
|
|
326
|
+
# Score the skill and print recommendations
|
|
332
327
|
asm eval ./my-skill
|
|
333
328
|
|
|
334
|
-
#
|
|
335
|
-
asm eval ./my-skill --
|
|
336
|
-
|
|
337
|
-
# Run the skillgrade runtime provider
|
|
338
|
-
asm eval ./my-skill --runtime --preset smoke
|
|
329
|
+
# CI-friendly machine-readable output
|
|
330
|
+
asm eval ./my-skill --machine
|
|
339
331
|
|
|
340
|
-
#
|
|
341
|
-
asm eval ./my-skill --
|
|
332
|
+
# Apply deterministic auto-fixes to SKILL.md
|
|
333
|
+
asm eval ./my-skill --fix
|
|
342
334
|
|
|
343
335
|
# List registered eval providers
|
|
344
336
|
asm eval-providers list
|
|
345
|
-
|
|
346
|
-
# Diff two provider versions before promoting an upgrade
|
|
347
|
-
asm eval ./my-skill --compare skillgrade@1.0.0,skillgrade@2.0.0-next
|
|
348
337
|
```
|
|
349
338
|
|
|
350
|
-
The eval surface is a pluggable provider framework: each provider implements a common `EvalProvider` contract and resolves via semver range
|
|
339
|
+
The eval surface is a pluggable provider framework: each provider implements a common `EvalProvider` contract and resolves via semver range. See [`docs/eval-providers.md`](./docs/eval-providers.md) for the provider model.
|
|
351
340
|
|
|
352
341
|
---
|
|
353
342
|
|
|
@@ -431,6 +420,28 @@ asm
|
|
|
431
420
|
<a href="#cli-commands"><strong>See All Commands →</strong></a>
|
|
432
421
|
</p>
|
|
433
422
|
|
|
423
|
+
### Pick one package manager
|
|
424
|
+
|
|
425
|
+
Install `asm` via **either** `npm` **or** `bun` — not both. Each global install drops an `asm` binary in a different directory (`/opt/homebrew/bin/asm` via npm, `~/.bun/bin/asm` via bun), and shells resolve whichever appears first on `PATH`. An older install can silently shadow a fresh upgrade: you run `asm --version` and still see the old version even though the upgrade succeeded.
|
|
426
|
+
|
|
427
|
+
<a id="troubleshooting"></a>
|
|
428
|
+
|
|
429
|
+
**Diagnose:** `asm --version` detects and warns when it sees multiple `asm` binaries on `PATH`. For a full report, run `asm doctor` — it lists the resolved path and any shadowed installs.
|
|
430
|
+
|
|
431
|
+
**Fix:** remove the duplicate install.
|
|
432
|
+
|
|
433
|
+
```bash
|
|
434
|
+
# If you're switching to npm:
|
|
435
|
+
bun remove -g agent-skill-manager
|
|
436
|
+
|
|
437
|
+
# If you're switching to bun:
|
|
438
|
+
npm uninstall -g agent-skill-manager
|
|
439
|
+
```
|
|
440
|
+
|
|
441
|
+
Re-run `asm --version` to confirm only one binary is left. The postinstall step emits the same warning during `npm install -g agent-skill-manager` so you catch it at install time.
|
|
442
|
+
|
|
443
|
+
> **Note:** Bun skips lifecycle scripts by default, so the postinstall warning does not fire for `bun add -g agent-skill-manager`. Use `asm --version` or `asm doctor` to check for shadowing after a bun install.
|
|
444
|
+
|
|
434
445
|
---
|
|
435
446
|
|
|
436
447
|
## Open-Source Skill Collections
|
|
@@ -514,7 +525,7 @@ Need a tool not listed? Add a custom provider entry to the config.
|
|
|
514
525
|
Yes. `asm` is MIT licensed and free forever. No accounts, no telemetry, no paywalls.
|
|
515
526
|
|
|
516
527
|
**Is it actively maintained?**
|
|
517
|
-
|
|
528
|
+
v2.0.0 shipped on April 19, 2026. The project has had 32 releases. Check the [changelog](docs/CHANGELOG.md) for the full history.
|
|
518
529
|
|
|
519
530
|
**Which AI agents does it support?**
|
|
520
531
|
18 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, Gemini CLI, Google Antigravity, Hermes, and a generic Agents provider. All 18 are enabled by default; disable any you don't need via `asm config edit`. You can also add any custom agent that stores skills as directories with a `SKILL.md` file.
|
|
@@ -578,8 +589,7 @@ asm
|
|
|
578
589
|
| `asm link <path> [<path2> ...]` | Symlink one or more local skills for live development |
|
|
579
590
|
| `asm audit` | Detect duplicate skills |
|
|
580
591
|
| `asm audit security <name>` | Run security audit on a skill |
|
|
581
|
-
| `asm eval <skill>` | Score a skill
|
|
582
|
-
| `asm eval <skill> --runtime` | Runtime evaluation via skillgrade (LLM-judge) |
|
|
592
|
+
| `asm eval <skill>` | Score a skill and print improvement suggestions |
|
|
583
593
|
| `asm eval-providers list` | List registered eval providers and versions |
|
|
584
594
|
| `asm stats` | Show aggregate skill metrics dashboard |
|
|
585
595
|
| `asm export` | Export skill inventory as JSON manifest |
|
|
@@ -648,12 +658,6 @@ Score a skill with the static quality provider:
|
|
|
648
658
|
asm eval ./my-skill
|
|
649
659
|
```
|
|
650
660
|
|
|
651
|
-
Run the skillgrade runtime evaluator (requires `skillgrade` on PATH):
|
|
652
|
-
|
|
653
|
-
```bash
|
|
654
|
-
asm eval ./my-skill --runtime --preset smoke
|
|
655
|
-
```
|
|
656
|
-
|
|
657
661
|
List registered eval providers:
|
|
658
662
|
|
|
659
663
|
```bash
|
|
@@ -1139,18 +1143,17 @@ agent-skill-manager/
|
|
|
1139
1143
|
<details>
|
|
1140
1144
|
<summary><strong>Documentation</strong></summary>
|
|
1141
1145
|
|
|
1142
|
-
| Document
|
|
1143
|
-
|
|
|
1144
|
-
| [Architecture](docs/ARCHITECTURE.md)
|
|
1145
|
-
| [Eval Providers](docs/eval-providers.md)
|
|
1146
|
-
| [
|
|
1147
|
-
| [
|
|
1148
|
-
| [
|
|
1149
|
-
| [
|
|
1150
|
-
| [
|
|
1151
|
-
| [
|
|
1152
|
-
| [
|
|
1153
|
-
| [Code of Conduct](CODE_OF_CONDUCT.md) | Community guidelines |
|
|
1146
|
+
| Document | Description |
|
|
1147
|
+
| ---------------------------------------- | -------------------------------------------------- |
|
|
1148
|
+
| [Architecture](docs/ARCHITECTURE.md) | System design, components, and data flow |
|
|
1149
|
+
| [Eval Providers](docs/eval-providers.md) | Pluggable eval framework and how to add a provider |
|
|
1150
|
+
| [Development](docs/DEVELOPMENT.md) | Local setup, testing, and debugging |
|
|
1151
|
+
| [Deployment](docs/DEPLOYMENT.md) | Publishing and CI pipeline |
|
|
1152
|
+
| [Changelog](docs/CHANGELOG.md) | Version history |
|
|
1153
|
+
| [Brand Kit](docs/brand_kit.md) | Logo, colors, and typography |
|
|
1154
|
+
| [Contributing](CONTRIBUTING.md) | How to contribute |
|
|
1155
|
+
| [Security](SECURITY.md) | Vulnerability reporting |
|
|
1156
|
+
| [Code of Conduct](CODE_OF_CONDUCT.md) | Community guidelines |
|
|
1154
1157
|
|
|
1155
1158
|
</details>
|
|
1156
1159
|
|