agent-skill-manager 1.14.0 → 1.15.1

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.
Files changed (32) hide show
  1. package/README.backup.md +38 -22
  2. package/README.md +76 -24
  3. package/data/skill-index/Affitor_affiliate-skills.json +97 -49
  4. package/data/skill-index/Eronred_aso-skills.json +61 -31
  5. package/data/skill-index/GPTomics_bioSkills.json +5097 -0
  6. package/data/skill-index/Galaxy-Dawn_claude-scholar.json +98 -51
  7. package/data/skill-index/K-Dense-AI_claude-scientific-skills.json +1977 -0
  8. package/data/skill-index/Leonxlnx_taste-skill.json +15 -8
  9. package/data/skill-index/Master-cai_Research-Paper-Writing-Skills.json +21 -0
  10. package/data/skill-index/MiniMax-AI_skills.json +34 -12
  11. package/data/skill-index/affaan-m_everything-claude-code.json +367 -184
  12. package/data/skill-index/anthropics_skills.json +37 -19
  13. package/data/skill-index/bytedance_deer-flow.json +213 -0
  14. package/data/skill-index/coreyhaines31_marketingskills.json +67 -34
  15. package/data/skill-index/github_awesome-copilot.json +2978 -0
  16. package/data/skill-index/himself65_finance-skills.json +17 -9
  17. package/data/skill-index/kemiljk_fluid-design.json +3 -2
  18. package/data/skill-index/kepano_obsidian-skills.json +11 -6
  19. package/data/skill-index/luongnv89_skills.json +73 -38
  20. package/data/skill-index/mattpocock_skills.json +35 -18
  21. package/data/skill-index/nextlevelbuilder_ui-ux-pro-max-skill.json +15 -8
  22. package/data/skill-index/obra_superpowers.json +29 -15
  23. package/data/skill-index/sickn33_antigravity-awesome-skills.json +2630 -1304
  24. package/data/skill-index/slavingia_skills.json +19 -10
  25. package/data/skill-index-resources.json +46 -19
  26. package/dist/agent-skill-manager.js +165 -164
  27. package/dist/{chunk-d8zyhd1t.js → chunk-6t9cpddh.js} +1 -1
  28. package/dist/{chunk-vhkxjexp.js → chunk-etgg42ff.js} +1 -1
  29. package/dist/{chunk-gvpf7w25.js → chunk-zzvkjwrx.js} +2 -2
  30. package/package.json +1 -1
  31. package/data/skill-index/agentskills_agentskills.json +0 -8
  32. package/data/skill-index/msitarzewski_agency-agents.json +0 -8
package/README.backup.md CHANGED
@@ -51,7 +51,7 @@ The more AI agents you use, the worse this gets. Every new tool adds another ski
51
51
  - **See everything at once** — List, search, and filter skills across all providers and scopes from one dashboard. No more `ls`-ing through hidden directories.
52
52
  - **Install from GitHub in one command** — `asm install github:user/repo` handles cloning, validation, and placement. Supports single-skill repos, multi-skill collections, subfolder URLs, and private repos via SSH.
53
53
  - **Catch problems before they bite** — Built-in security scanning flags dangerous patterns (shell execution, network access, credential exposure, obfuscation) before you install. Duplicate audit finds and cleans redundant skills across providers.
54
- - **Works with every major agent** — 15 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, and a generic Agents provider. Add custom providers in seconds via config.
54
+ - **Works with every major agent** — 17 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, Gemini CLI, Google Antigravity, and a generic Agents provider. Add custom providers in seconds via config.
55
55
  - **Two interfaces, one tool** — Full interactive TUI with keyboard navigation, search, and detail views. Or use the CLI with `--json` for scripting and automation.
56
56
 
57
57
  <p align="center">
@@ -130,25 +130,27 @@ asm install github:anthropics/skills --all # install everything
130
130
 
131
131
  ## Supported Agent Tools
132
132
 
133
- `asm` ships with **15 built-in providers**. The first 4 are enabled by default; the rest can be enabled via `asm config edit`.
134
-
135
- | Tool | Global Path | Project Path | Default |
136
- | ---------------- | --------------------------------- | ----------------------- | :------: |
137
- | Claude Code | `~/.claude/skills/` | `.claude/skills/` | enabled |
138
- | Codex | `~/.codex/skills/` | `.codex/skills/` | enabled |
139
- | OpenClaw | `~/.openclaw/skills/` | `.openclaw/skills/` | enabled |
140
- | Agents (generic) | `~/.agents/skills/` | `.agents/skills/` | enabled |
141
- | Cursor | `~/.cursor/rules/` | `.cursor/rules/` | disabled |
142
- | Windsurf | `~/.windsurf/rules/` | `.windsurf/rules/` | disabled |
143
- | Cline | `~/Documents/Cline/Rules/` | `.clinerules/` | disabled |
144
- | Roo Code | `~/.roo/rules/` | `.roo/rules/` | disabled |
145
- | Continue | `~/.continue/rules/` | `.continue/rules/` | disabled |
146
- | GitHub Copilot | `~/.github/instructions/` | `.github/instructions/` | disabled |
147
- | Aider | `~/.aider/skills/` | `.aider/skills/` | disabled |
148
- | OpenCode | `~/.config/opencode/skills/` | `.opencode/skills/` | disabled |
149
- | Zed | `~/.config/zed/prompt_overrides/` | `.zed/rules/` | disabled |
150
- | Augment | `~/.augment/rules/` | `.augment/rules/` | disabled |
151
- | Amp | `~/.amp/skills/` | `.amp/skills/` | disabled |
133
+ `asm` ships with **17 built-in providers**. The first 4 are enabled by default; the rest can be enabled via `asm config edit`.
134
+
135
+ | Tool | Global Path | Project Path | Default |
136
+ | ------------------ | --------------------------------- | ----------------------- | :------: |
137
+ | Claude Code | `~/.claude/skills/` | `.claude/skills/` | enabled |
138
+ | Codex | `~/.codex/skills/` | `.codex/skills/` | enabled |
139
+ | OpenClaw | `~/.openclaw/skills/` | `.openclaw/skills/` | enabled |
140
+ | Agents (generic) | `~/.agents/skills/` | `.agents/skills/` | enabled |
141
+ | Cursor | `~/.cursor/rules/` | `.cursor/rules/` | disabled |
142
+ | Windsurf | `~/.windsurf/rules/` | `.windsurf/rules/` | disabled |
143
+ | Cline | `~/Documents/Cline/Rules/` | `.clinerules/` | disabled |
144
+ | Roo Code | `~/.roo/rules/` | `.roo/rules/` | disabled |
145
+ | Continue | `~/.continue/rules/` | `.continue/rules/` | disabled |
146
+ | GitHub Copilot | `~/.github/instructions/` | `.github/instructions/` | disabled |
147
+ | Aider | `~/.aider/skills/` | `.aider/skills/` | disabled |
148
+ | OpenCode | `~/.config/opencode/skills/` | `.opencode/skills/` | disabled |
149
+ | Zed | `~/.config/zed/prompt_overrides/` | `.zed/rules/` | disabled |
150
+ | Augment | `~/.augment/rules/` | `.augment/rules/` | disabled |
151
+ | Amp | `~/.amp/skills/` | `.amp/skills/` | disabled |
152
+ | Gemini CLI | `~/.gemini/skills/` | `.gemini/skills/` | disabled |
153
+ | Google Antigravity | `~/.antigravity/skills/` | `.antigravity/skills/` | disabled |
152
154
 
153
155
  Enable a provider:
154
156
 
@@ -169,7 +171,7 @@ Yes. `asm` is MIT licensed and free forever. No accounts, no telemetry, no paywa
169
171
  v1.10.0 shipped on March 18, 2026. The project has had 10 releases in the past week. Check the [changelog](docs/CHANGELOG.md) for the full history.
170
172
 
171
173
  **Which AI agents does it support?**
172
- 15 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, and a generic Agents provider. The first 4 are enabled by default; enable the rest via `asm config edit`. You can also add any custom agent that stores skills as directories with a `SKILL.md` file.
174
+ 17 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, Gemini CLI, Google Antigravity, and a generic Agents provider. The first 4 are enabled by default; enable the rest via `asm config edit`. You can also add any custom agent that stores skills as directories with a `SKILL.md` file.
173
175
 
174
176
  **How does it compare to managing skills manually?**
175
177
  Manual management means remembering where each agent stores skills, cloning repos by hand, checking for duplicates yourself, and having no security scanning. `asm` automates all of that with one command.
@@ -342,7 +344,7 @@ The install command clones the repository, validates `SKILL.md` files, scans for
342
344
  <details>
343
345
  <summary><strong>Configuration</strong></summary>
344
346
 
345
- On first run, a config file is created at `~/.config/agent-skill-manager/config.json` with 15 default providers. The first 4 are enabled; the rest are disabled until you enable them:
347
+ On first run, a config file is created at `~/.config/agent-skill-manager/config.json` with 17 default providers. The first 4 are enabled; the rest are disabled until you enable them:
346
348
 
347
349
  ```json
348
350
  {
@@ -452,6 +454,20 @@ On first run, a config file is created at `~/.config/agent-skill-manager/config.
452
454
  "global": "~/.amp/skills",
453
455
  "project": ".amp/skills",
454
456
  "enabled": false
457
+ },
458
+ {
459
+ "name": "gemini",
460
+ "label": "Gemini CLI",
461
+ "global": "~/.gemini/skills",
462
+ "project": ".gemini/skills",
463
+ "enabled": false
464
+ },
465
+ {
466
+ "name": "antigravity",
467
+ "label": "Google Antigravity",
468
+ "global": "~/.antigravity/skills",
469
+ "project": ".antigravity/skills",
470
+ "enabled": false
455
471
  }
456
472
  ],
457
473
  "customPaths": [],
package/README.md CHANGED
@@ -25,7 +25,7 @@
25
25
  <p align="center">
26
26
  <a href="#get-started-in-30-seconds"><strong>Get Started in 30 Seconds &rarr;</strong></a>
27
27
  &nbsp;&nbsp;|&nbsp;&nbsp;
28
- <a href="https://luongnv.com/asm/"><strong>Browse 1,700+ Skills Online &rarr;</strong></a>
28
+ <a href="https://luongnv.com/asm/"><strong>Browse 2,600+ Skills Online &rarr;</strong></a>
29
29
  </p>
30
30
 
31
31
  ---
@@ -62,7 +62,7 @@ The more AI agents you use, the worse this gets. Every new tool adds another ski
62
62
  - **Install from GitHub in one command** — `asm install github:user/repo` handles cloning, validation, and placement. Supports single-skill repos, multi-skill collections, subfolder URLs, and private repos via SSH.
63
63
  - **Catch problems before they bite** — Built-in security scanning flags dangerous patterns (shell execution, network access, credential exposure, obfuscation) before you install. Duplicate audit finds and cleans redundant skills across providers.
64
64
  - **Create and test skills locally** — Scaffold new skills with `asm init`, symlink them for live development with `asm link`, audit for security issues, and verify metadata — all before publishing. [See the full local dev workflow &darr;](#build-test-and-ship-your-own-skills)
65
- - **Works with every major agent** — 15 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, and a generic Agents provider. Add custom providers in seconds via config.
65
+ - **Works with every major agent** — 17 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, Gemini CLI, Google Antigravity, and a generic Agents provider. Add custom providers in seconds via config.
66
66
  - **Two interfaces, one tool** — Full interactive TUI with keyboard navigation, search, and detail views. Or use the CLI with `--json` for scripting and automation.
67
67
 
68
68
  <p align="center">
@@ -217,6 +217,42 @@ Whether you're building skills for yourself or publishing them for the community
217
217
 
218
218
  ---
219
219
 
220
+ ## Skill Verification
221
+
222
+ Skills indexed by `asm` are automatically evaluated against a set of verification criteria. Skills that pass all criteria receive a **verified** badge in the catalog and `"verified": true` in the index JSON. Skills that fail any criterion are still indexed but marked as unverified.
223
+
224
+ ### Verification Criteria
225
+
226
+ A skill must satisfy **all four** of the following to be verified:
227
+
228
+ 1. **Valid frontmatter** -- The SKILL.md file must contain YAML frontmatter with both a `name` and a `description` field. Empty or whitespace-only values fail this check.
229
+
230
+ 2. **Meaningful body content** -- The markdown body (everything after the frontmatter block) must contain at least 20 characters of instruction text. A SKILL.md that is only frontmatter with no real guidance for the agent will fail.
231
+
232
+ 3. **No malicious patterns** -- The full SKILL.md content is scanned for dangerous code patterns:
233
+ - `atob()` calls (runtime base64 decoding / obfuscation)
234
+ - Suspicious base64-encoded strings (40+ character base64 blocks with padding)
235
+ - Hex-escape sequences (4+ consecutive `\xNN` escapes)
236
+ - Hardcoded credentials (`API_KEY`, `SECRET_KEY`, or `PASSWORD` assignments)
237
+
238
+ 4. **Proper structure** -- The skill directory must exist and contain a `SKILL.md` file that the ingestion pipeline can read.
239
+
240
+ ### How to Reproduce Locally
241
+
242
+ You can verify your skill before publishing:
243
+
244
+ ```bash
245
+ # Index your repo -- verification runs automatically during ingestion
246
+ asm index ingest github:your-user/your-repo
247
+
248
+ # Check the output JSON for the verified field
249
+ asm index search "your-skill" --json
250
+ ```
251
+
252
+ 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.
253
+
254
+ ---
255
+
220
256
  ## Get Started in 30 Seconds
221
257
 
222
258
  ### npm (recommended)
@@ -247,7 +283,7 @@ asm
247
283
 
248
284
  ## Open-Source Skill Collections
249
285
 
250
- A curated list of skill repositories you can install with a single command. Over **1,500 skills** available across these collections:
286
+ A curated list of skill repositories you can install with a single command. Over **2,600 skills** available across these collections:
251
287
 
252
288
  > **Last updated:** 2026-03-18
253
289
 
@@ -286,25 +322,27 @@ asm install github:anthropics/skills --all
286
322
 
287
323
  ## Supported Agent Tools
288
324
 
289
- `asm` ships with **15 built-in providers**, all enabled by default. Disable any you don't need via `asm config edit`.
290
-
291
- | Tool | Global Path | Project Path | Default |
292
- | ---------------- | --------------------------------- | ----------------------- | :-----: |
293
- | Claude Code | `~/.claude/skills/` | `.claude/skills/` | enabled |
294
- | Codex | `~/.codex/skills/` | `.codex/skills/` | enabled |
295
- | OpenClaw | `~/.openclaw/skills/` | `.openclaw/skills/` | enabled |
296
- | Agents (generic) | `~/.agents/skills/` | `.agents/skills/` | enabled |
297
- | Cursor | `~/.cursor/rules/` | `.cursor/rules/` | enabled |
298
- | Windsurf | `~/.windsurf/rules/` | `.windsurf/rules/` | enabled |
299
- | Cline | `~/Documents/Cline/Rules/` | `.clinerules/` | enabled |
300
- | Roo Code | `~/.roo/rules/` | `.roo/rules/` | enabled |
301
- | Continue | `~/.continue/rules/` | `.continue/rules/` | enabled |
302
- | GitHub Copilot | `~/.github/instructions/` | `.github/instructions/` | enabled |
303
- | Aider | `~/.aider/skills/` | `.aider/skills/` | enabled |
304
- | OpenCode | `~/.config/opencode/skills/` | `.opencode/skills/` | enabled |
305
- | Zed | `~/.config/zed/prompt_overrides/` | `.zed/rules/` | enabled |
306
- | Augment | `~/.augment/rules/` | `.augment/rules/` | enabled |
307
- | Amp | `~/.amp/skills/` | `.amp/skills/` | enabled |
325
+ `asm` ships with **17 built-in providers**, all enabled by default. Disable any you don't need via `asm config edit`.
326
+
327
+ | Tool | Global Path | Project Path | Default |
328
+ | ------------------ | --------------------------------- | ----------------------- | :-----: |
329
+ | Claude Code | `~/.claude/skills/` | `.claude/skills/` | enabled |
330
+ | Codex | `~/.codex/skills/` | `.codex/skills/` | enabled |
331
+ | OpenClaw | `~/.openclaw/skills/` | `.openclaw/skills/` | enabled |
332
+ | Agents (generic) | `~/.agents/skills/` | `.agents/skills/` | enabled |
333
+ | Cursor | `~/.cursor/rules/` | `.cursor/rules/` | enabled |
334
+ | Windsurf | `~/.windsurf/rules/` | `.windsurf/rules/` | enabled |
335
+ | Cline | `~/Documents/Cline/Rules/` | `.clinerules/` | enabled |
336
+ | Roo Code | `~/.roo/rules/` | `.roo/rules/` | enabled |
337
+ | Continue | `~/.continue/rules/` | `.continue/rules/` | enabled |
338
+ | GitHub Copilot | `~/.github/instructions/` | `.github/instructions/` | enabled |
339
+ | Aider | `~/.aider/skills/` | `.aider/skills/` | enabled |
340
+ | OpenCode | `~/.config/opencode/skills/` | `.opencode/skills/` | enabled |
341
+ | Zed | `~/.config/zed/prompt_overrides/` | `.zed/rules/` | enabled |
342
+ | Augment | `~/.augment/rules/` | `.augment/rules/` | enabled |
343
+ | Amp | `~/.amp/skills/` | `.amp/skills/` | enabled |
344
+ | Gemini CLI | `~/.gemini/skills/` | `.gemini/skills/` | enabled |
345
+ | Google Antigravity | `~/.antigravity/skills/` | `.antigravity/skills/` | enabled |
308
346
 
309
347
  Disable a provider — opens config in `$EDITOR`, set `"enabled": false` for any provider:
310
348
 
@@ -325,7 +363,7 @@ Yes. `asm` is MIT licensed and free forever. No accounts, no telemetry, no paywa
325
363
  v1.12.0 shipped on March 23, 2026. The project has had 12 releases in the past week. Check the [changelog](docs/CHANGELOG.md) for the full history.
326
364
 
327
365
  **Which AI agents does it support?**
328
- 15 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, and a generic Agents provider. All 15 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.
366
+ 17 providers built-in: Claude Code, Codex, OpenClaw, Cursor, Windsurf, Cline, Roo Code, Continue, GitHub Copilot, Aider, OpenCode, Zed, Augment, Amp, Gemini CLI, Google Antigravity, and a generic Agents provider. All 17 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.
329
367
 
330
368
  **How does it compare to managing skills manually?**
331
369
  Manual management means remembering where each agent stores skills, cloning repos by hand, checking for duplicates yourself, and having no security scanning. `asm` automates all of that with one command.
@@ -605,7 +643,7 @@ The install command clones the repository, validates `SKILL.md` files, scans for
605
643
  <details>
606
644
  <summary><strong>Configuration</strong></summary>
607
645
 
608
- On first run, a config file is created at `~/.config/agent-skill-manager/config.json` with 15 default providers, all enabled:
646
+ On first run, a config file is created at `~/.config/agent-skill-manager/config.json` with 17 default providers, all enabled:
609
647
 
610
648
  ```json
611
649
  {
@@ -715,6 +753,20 @@ On first run, a config file is created at `~/.config/agent-skill-manager/config.
715
753
  "global": "~/.amp/skills",
716
754
  "project": ".amp/skills",
717
755
  "enabled": false
756
+ },
757
+ {
758
+ "name": "gemini",
759
+ "label": "Gemini CLI",
760
+ "global": "~/.gemini/skills",
761
+ "project": ".gemini/skills",
762
+ "enabled": false
763
+ },
764
+ {
765
+ "name": "antigravity",
766
+ "label": "Google Antigravity",
767
+ "global": "~/.antigravity/skills",
768
+ "project": ".antigravity/skills",
769
+ "enabled": false
718
770
  }
719
771
  ],
720
772
  "customPaths": [],