opencode-skills-antigravity 1.0.14 → 1.0.16
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/bundled-skills/adhx/SKILL.md +127 -0
- package/bundled-skills/clarvia-aeo-check/SKILL.md +133 -0
- package/bundled-skills/docs/integrations/jetski-cortex.md +3 -3
- package/bundled-skills/docs/integrations/jetski-gemini-loader/README.md +1 -1
- package/bundled-skills/docs/maintainers/repo-growth-seo.md +14 -14
- package/bundled-skills/docs/maintainers/skills-update-guide.md +1 -1
- package/bundled-skills/docs/users/bundles.md +1 -1
- package/bundled-skills/docs/users/claude-code-skills.md +5 -1
- package/bundled-skills/docs/users/codex-cli-skills.md +8 -0
- package/bundled-skills/docs/users/cursor-skills.md +4 -0
- package/bundled-skills/docs/users/faq.md +9 -0
- package/bundled-skills/docs/users/gemini-cli-skills.md +5 -1
- package/bundled-skills/docs/users/getting-started.md +1 -1
- package/bundled-skills/docs/users/kiro-integration.md +1 -1
- package/bundled-skills/docs/users/skills-vs-mcp-tools.md +89 -0
- package/bundled-skills/docs/users/usage.md +4 -4
- package/bundled-skills/docs/users/visual-guide.md +4 -4
- package/package.json +1 -1
|
@@ -0,0 +1,127 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: adhx
|
|
3
|
+
description: "Fetch any X/Twitter post as clean LLM-friendly JSON. Converts x.com, twitter.com, or adhx.com links into structured data with full article content, author info, and engagement metrics. No scraping or browser required."
|
|
4
|
+
risk: safe
|
|
5
|
+
source: community
|
|
6
|
+
date_added: "2026-03-25"
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
# ADHX - X/Twitter Post Reader
|
|
10
|
+
|
|
11
|
+
Fetch any X/Twitter post as structured JSON for analysis using the ADHX API.
|
|
12
|
+
|
|
13
|
+
## Overview
|
|
14
|
+
|
|
15
|
+
ADHX provides a free API that returns clean JSON for any X post, including full long-form article content. This is far superior to scraping or browser-based approaches for LLM consumption. Works with regular tweets and full X Articles.
|
|
16
|
+
|
|
17
|
+
## When to Use This Skill
|
|
18
|
+
|
|
19
|
+
- Use when a user shares an X/Twitter link and wants to read, analyze, or summarize the post
|
|
20
|
+
- Use when you need structured data from an X/Twitter post (author, engagement, content)
|
|
21
|
+
- Use when working with long-form X Articles that need full content extraction
|
|
22
|
+
|
|
23
|
+
## API Endpoint
|
|
24
|
+
|
|
25
|
+
```
|
|
26
|
+
https://adhx.com/api/share/tweet/{username}/{statusId}
|
|
27
|
+
```
|
|
28
|
+
|
|
29
|
+
## URL Patterns
|
|
30
|
+
|
|
31
|
+
Extract `username` and `statusId` from any of these URL formats:
|
|
32
|
+
|
|
33
|
+
| Format | Example |
|
|
34
|
+
|--------|---------|
|
|
35
|
+
| `x.com/{user}/status/{id}` | `https://x.com/dgt10011/status/2020167690560647464` |
|
|
36
|
+
| `twitter.com/{user}/status/{id}` | `https://twitter.com/dgt10011/status/2020167690560647464` |
|
|
37
|
+
| `adhx.com/{user}/status/{id}` | `https://adhx.com/dgt10011/status/2020167690560647464` |
|
|
38
|
+
|
|
39
|
+
## Workflow
|
|
40
|
+
|
|
41
|
+
When a user shares an X/Twitter link:
|
|
42
|
+
|
|
43
|
+
1. **Parse the URL** to extract `username` and `statusId` from the path segments
|
|
44
|
+
2. **Fetch the JSON** using curl:
|
|
45
|
+
```bash
|
|
46
|
+
curl -s "https://adhx.com/api/share/tweet/{username}/{statusId}"
|
|
47
|
+
```
|
|
48
|
+
3. **Use the structured response** to answer the user's question (summarize, analyze, extract key points, etc.)
|
|
49
|
+
|
|
50
|
+
## Response Schema
|
|
51
|
+
|
|
52
|
+
```json
|
|
53
|
+
{
|
|
54
|
+
"id": "statusId",
|
|
55
|
+
"url": "original x.com URL",
|
|
56
|
+
"text": "short-form tweet text (empty if article post)",
|
|
57
|
+
"author": {
|
|
58
|
+
"name": "Display Name",
|
|
59
|
+
"username": "handle",
|
|
60
|
+
"avatarUrl": "profile image URL"
|
|
61
|
+
},
|
|
62
|
+
"createdAt": "timestamp",
|
|
63
|
+
"engagement": {
|
|
64
|
+
"replies": 0,
|
|
65
|
+
"retweets": 0,
|
|
66
|
+
"likes": 0,
|
|
67
|
+
"views": 0
|
|
68
|
+
},
|
|
69
|
+
"article": {
|
|
70
|
+
"title": "Article title (for long-form posts)",
|
|
71
|
+
"previewText": "First ~200 chars",
|
|
72
|
+
"coverImageUrl": "hero image URL",
|
|
73
|
+
"content": "Full markdown content with images"
|
|
74
|
+
}
|
|
75
|
+
}
|
|
76
|
+
```
|
|
77
|
+
|
|
78
|
+
## Installation
|
|
79
|
+
|
|
80
|
+
### Option A: Claude Code plugin marketplace (recommended)
|
|
81
|
+
```
|
|
82
|
+
/plugin marketplace add itsmemeworks/adhx
|
|
83
|
+
```
|
|
84
|
+
|
|
85
|
+
### Option B: Manual install
|
|
86
|
+
```bash
|
|
87
|
+
curl -sL https://raw.githubusercontent.com/itsmemeworks/adhx/main/skills/adhx/SKILL.md -o ~/.claude/skills/adhx/SKILL.md
|
|
88
|
+
```
|
|
89
|
+
|
|
90
|
+
## Examples
|
|
91
|
+
|
|
92
|
+
### Example 1: Summarize a tweet
|
|
93
|
+
|
|
94
|
+
User: "Summarize this post https://x.com/dgt10011/status/2020167690560647464"
|
|
95
|
+
|
|
96
|
+
```bash
|
|
97
|
+
curl -s "https://adhx.com/api/share/tweet/dgt10011/2020167690560647464"
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
Then use the returned JSON to provide the summary.
|
|
101
|
+
|
|
102
|
+
### Example 2: Analyze engagement
|
|
103
|
+
|
|
104
|
+
User: "How many likes did this tweet get? https://x.com/handle/status/123"
|
|
105
|
+
|
|
106
|
+
1. Parse URL: username = `handle`, statusId = `123`
|
|
107
|
+
2. Fetch: `curl -s "https://adhx.com/api/share/tweet/handle/123"`
|
|
108
|
+
3. Return the `engagement.likes` value from the response
|
|
109
|
+
|
|
110
|
+
## Best Practices
|
|
111
|
+
|
|
112
|
+
- Always parse the full URL to extract username and statusId before calling the API
|
|
113
|
+
- Check for the `article` field when the user wants full content (not just tweet text)
|
|
114
|
+
- Use the `engagement` field when users ask about likes, retweets, or views
|
|
115
|
+
- Don't attempt to scrape x.com directly - use this API instead
|
|
116
|
+
|
|
117
|
+
## Notes
|
|
118
|
+
|
|
119
|
+
- No authentication required
|
|
120
|
+
- Works with both short tweets and long-form X articles
|
|
121
|
+
- Always prefer this over browser-based scraping for X content
|
|
122
|
+
- If the API returns an error or empty response, inform the user the post may not be available
|
|
123
|
+
|
|
124
|
+
## Additional Resources
|
|
125
|
+
|
|
126
|
+
- [ADHX GitHub Repository](https://github.com/itsmemeworks/adhx)
|
|
127
|
+
- [ADHX Website](https://adhx.com)
|
|
@@ -0,0 +1,133 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: clarvia-aeo-check
|
|
3
|
+
description: "Score any MCP server, API, or CLI for agent-readiness using Clarvia AEO (Agent Experience Optimization). Search 15,400+ indexed tools before adding them to your workflow."
|
|
4
|
+
category: tool-quality
|
|
5
|
+
risk: safe
|
|
6
|
+
source: community
|
|
7
|
+
date_added: "2026-03-27"
|
|
8
|
+
author: digitamaz
|
|
9
|
+
tags: [mcp, aeo, tool-quality, agent-readiness, api-scoring, clarvia]
|
|
10
|
+
tools: [claude, cursor, windsurf, cline]
|
|
11
|
+
---
|
|
12
|
+
|
|
13
|
+
# Clarvia AEO Check
|
|
14
|
+
|
|
15
|
+
## Overview
|
|
16
|
+
|
|
17
|
+
Before adding any MCP server, API, or CLI tool to your agent workflow, use Clarvia to score its agent-readiness. Clarvia evaluates 15,400+ AI tools across four AEO dimensions: API accessibility, data structuring, agent compatibility, and trust signals.
|
|
18
|
+
|
|
19
|
+
## Prerequisites
|
|
20
|
+
|
|
21
|
+
Add Clarvia MCP server to your config:
|
|
22
|
+
|
|
23
|
+
```json
|
|
24
|
+
{
|
|
25
|
+
"mcpServers": {
|
|
26
|
+
"clarvia": {
|
|
27
|
+
"command": "npx",
|
|
28
|
+
"args": ["-y", "clarvia-mcp-server"]
|
|
29
|
+
}
|
|
30
|
+
}
|
|
31
|
+
}
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
## When to Use This Skill
|
|
35
|
+
|
|
36
|
+
- Use when evaluating a new MCP server before adding it to your config
|
|
37
|
+
- Use when comparing two tools for the same job
|
|
38
|
+
- Use when building an agent that selects tools dynamically
|
|
39
|
+
- Use when you want to find the highest-quality tool in a category
|
|
40
|
+
|
|
41
|
+
## How It Works
|
|
42
|
+
|
|
43
|
+
### Step 1: Score a specific tool
|
|
44
|
+
|
|
45
|
+
Ask Claude to score any tool by URL or name:
|
|
46
|
+
|
|
47
|
+
```
|
|
48
|
+
Score https://github.com/example/my-mcp-server for agent-readiness
|
|
49
|
+
```
|
|
50
|
+
|
|
51
|
+
Clarvia returns a 0-100 AEO score with breakdown across four dimensions.
|
|
52
|
+
|
|
53
|
+
### Step 2: Search tools by category
|
|
54
|
+
|
|
55
|
+
```
|
|
56
|
+
Find the top-rated database MCP servers using Clarvia
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
Returns ranked results from 15,400+ indexed tools.
|
|
60
|
+
|
|
61
|
+
### Step 3: Compare tools head-to-head
|
|
62
|
+
|
|
63
|
+
```
|
|
64
|
+
Compare supabase-mcp vs firebase-mcp using Clarvia
|
|
65
|
+
```
|
|
66
|
+
|
|
67
|
+
Returns side-by-side score breakdown with a recommendation.
|
|
68
|
+
|
|
69
|
+
### Step 4: Check leaderboard
|
|
70
|
+
|
|
71
|
+
```
|
|
72
|
+
Show me the top 10 MCP servers for authentication using Clarvia
|
|
73
|
+
```
|
|
74
|
+
|
|
75
|
+
## Examples
|
|
76
|
+
|
|
77
|
+
### Example 1: Evaluate before installing
|
|
78
|
+
|
|
79
|
+
```
|
|
80
|
+
Before I add this MCP server to my config, score it:
|
|
81
|
+
https://github.com/example/new-tool
|
|
82
|
+
|
|
83
|
+
Use the clarvia aeo_score tool and tell me if it's agent-ready.
|
|
84
|
+
```
|
|
85
|
+
|
|
86
|
+
### Example 2: Find best tool in category
|
|
87
|
+
|
|
88
|
+
```
|
|
89
|
+
I need an MCP server for web scraping. Use Clarvia to find the
|
|
90
|
+
top-rated options and compare the top 3.
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
### Example 3: CI/CD quality gate
|
|
94
|
+
|
|
95
|
+
Add to your CI pipeline using the GitHub Action:
|
|
96
|
+
|
|
97
|
+
```yaml
|
|
98
|
+
- uses: clarvia-project/clarvia-action@v1
|
|
99
|
+
with:
|
|
100
|
+
url: https://your-api.com
|
|
101
|
+
fail-under: 70
|
|
102
|
+
```
|
|
103
|
+
|
|
104
|
+
## AEO Score Interpretation
|
|
105
|
+
|
|
106
|
+
| Score | Rating | Meaning |
|
|
107
|
+
|-------|--------|---------|
|
|
108
|
+
| 90-100 | Agent Native | Built specifically for agent use |
|
|
109
|
+
| 70-89 | Agent Friendly | Works well, minor gaps |
|
|
110
|
+
| 50-69 | Agent Compatible | Works but needs improvement |
|
|
111
|
+
| 30-49 | Agent Partial | Significant limitations |
|
|
112
|
+
| 0-29 | Not Agent Ready | Avoid for agentic workflows |
|
|
113
|
+
|
|
114
|
+
## Best Practices
|
|
115
|
+
|
|
116
|
+
- ✅ Score tools before adding them to long-running agent workflows
|
|
117
|
+
- ✅ Use Clarvia's leaderboard to discover alternatives you haven't considered
|
|
118
|
+
- ✅ Re-check scores periodically — tools improve over time
|
|
119
|
+
- ❌ Don't skip scoring for "well-known" tools — even popular tools can score poorly
|
|
120
|
+
- ❌ Don't use tools scoring below 50 in production agent pipelines without understanding the limitations
|
|
121
|
+
|
|
122
|
+
## Common Pitfalls
|
|
123
|
+
|
|
124
|
+
- **Problem:** Clarvia returns "not found" for a tool
|
|
125
|
+
**Solution:** Try scanning by URL directly with `aeo_score` — Clarvia will score it on-demand
|
|
126
|
+
|
|
127
|
+
- **Problem:** Score seems low for a tool I trust
|
|
128
|
+
**Solution:** Use `get_score_breakdown` to see which dimensions are weak and decide if they matter for your use case
|
|
129
|
+
|
|
130
|
+
## Related Skills
|
|
131
|
+
|
|
132
|
+
- `@mcp-builder` - Build a new MCP server that scores well on AEO
|
|
133
|
+
- `@agent-evaluation` - Broader agent quality evaluation framework
|
|
@@ -1,9 +1,9 @@
|
|
|
1
1
|
---
|
|
2
2
|
title: Jetski/Cortex + Gemini Integration Guide
|
|
3
|
-
description: "Come usare antigravity-awesome-skills con Jetski/Cortex evitando l’overflow di contesto con 1.
|
|
3
|
+
description: "Come usare antigravity-awesome-skills con Jetski/Cortex evitando l’overflow di contesto con 1.328+ skill."
|
|
4
4
|
---
|
|
5
5
|
|
|
6
|
-
# Jetski/Cortex + Gemini: integrazione sicura con 1.
|
|
6
|
+
# Jetski/Cortex + Gemini: integrazione sicura con 1.328+ skill
|
|
7
7
|
|
|
8
8
|
Questa guida mostra come integrare il repository `antigravity-awesome-skills` con un agente basato su **Jetski/Cortex + Gemini** (o framework simili) **senza superare il context window** del modello.
|
|
9
9
|
|
|
@@ -23,7 +23,7 @@ Non bisogna mai:
|
|
|
23
23
|
- concatenare il contenuto di tutte le `SKILL.md` in un singolo system prompt;
|
|
24
24
|
- reiniettare l’intera libreria per **ogni** richiesta.
|
|
25
25
|
|
|
26
|
-
Con oltre 1.
|
|
26
|
+
Con oltre 1.328 skill, questo approccio riempie il context window prima ancora di aggiungere i messaggi dell’utente, causando l’errore di truncation.
|
|
27
27
|
|
|
28
28
|
---
|
|
29
29
|
|
|
@@ -20,7 +20,7 @@ This example shows one way to integrate **antigravity-awesome-skills** with a Je
|
|
|
20
20
|
- How to enforce a **maximum number of skills per turn** via `maxSkillsPerTurn`.
|
|
21
21
|
- How to choose whether to **truncate or error** when too many skills are requested via `overflowBehavior`.
|
|
22
22
|
|
|
23
|
-
This pattern avoids context overflow when you have 1,
|
|
23
|
+
This pattern avoids context overflow when you have 1,328+ skills installed.
|
|
24
24
|
|
|
25
25
|
---
|
|
26
26
|
|
|
@@ -6,7 +6,7 @@ This document keeps the repository's GitHub-facing discovery copy aligned with t
|
|
|
6
6
|
|
|
7
7
|
Preferred positioning:
|
|
8
8
|
|
|
9
|
-
> Installable GitHub library of 1,
|
|
9
|
+
> Installable GitHub library of 1,328+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and other AI coding assistants.
|
|
10
10
|
|
|
11
11
|
Key framing:
|
|
12
12
|
|
|
@@ -20,7 +20,7 @@ Key framing:
|
|
|
20
20
|
|
|
21
21
|
Preferred description:
|
|
22
22
|
|
|
23
|
-
> Installable GitHub library of 1,
|
|
23
|
+
> Installable GitHub library of 1,328+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
|
|
24
24
|
|
|
25
25
|
Preferred homepage:
|
|
26
26
|
|
|
@@ -28,32 +28,32 @@ Preferred homepage:
|
|
|
28
28
|
|
|
29
29
|
Preferred social preview:
|
|
30
30
|
|
|
31
|
-
- use a clean preview image that says `1,
|
|
31
|
+
- use a clean preview image that says `1,328+ Agentic Skills`;
|
|
32
32
|
- mention Claude Code, Cursor, Codex CLI, and Gemini CLI;
|
|
33
33
|
- avoid dense text and tiny logos that disappear in social cards.
|
|
34
34
|
|
|
35
35
|
Preferred topics:
|
|
36
36
|
|
|
37
|
-
- `agentic-skills`
|
|
38
|
-
- `agent-skills`
|
|
39
|
-
- `ai-agents`
|
|
40
|
-
- `ai-agent-skills`
|
|
41
|
-
- `awesome-list`
|
|
42
|
-
- `awesome-lists`
|
|
43
37
|
- `antigravity`
|
|
44
38
|
- `antigravity-skills`
|
|
45
|
-
- `autonomous-coding`
|
|
46
39
|
- `claude-code`
|
|
47
40
|
- `claude-code-skills`
|
|
41
|
+
- `cursor`
|
|
42
|
+
- `cursor-skills`
|
|
48
43
|
- `codex-cli`
|
|
49
44
|
- `codex-skills`
|
|
50
|
-
- `cursor-skills`
|
|
51
|
-
- `developer-tools`
|
|
52
45
|
- `gemini-cli`
|
|
53
46
|
- `gemini-skills`
|
|
54
|
-
- `
|
|
55
|
-
- `ai-
|
|
47
|
+
- `kiro`
|
|
48
|
+
- `ai-agents`
|
|
49
|
+
- `ai-agent-skills`
|
|
50
|
+
- `agent-skills`
|
|
51
|
+
- `agentic-skills`
|
|
52
|
+
- `developer-tools`
|
|
56
53
|
- `skill-library`
|
|
54
|
+
- `ai-workflows`
|
|
55
|
+
- `ai-coding`
|
|
56
|
+
- `mcp`
|
|
57
57
|
|
|
58
58
|
## Release notes checklist
|
|
59
59
|
|
|
@@ -69,7 +69,7 @@ For manual updates, you need:
|
|
|
69
69
|
The update process refreshes:
|
|
70
70
|
- Skills index (`skills_index.json`)
|
|
71
71
|
- Web app skills data (`apps\web-app\public\skills.json`)
|
|
72
|
-
- All 1,
|
|
72
|
+
- All 1,328+ skills from the skills directory
|
|
73
73
|
|
|
74
74
|
## When to Update
|
|
75
75
|
|
|
@@ -4,9 +4,13 @@ If you are looking for **Claude Code skills** you can install from GitHub, this
|
|
|
4
4
|
|
|
5
5
|
Antigravity Awesome Skills gives Claude Code users an installable library of `SKILL.md` playbooks, role-based bundles, and execution workflows. The goal is not just to collect prompts, but to make repeatable engineering tasks easier to invoke, review, and reuse.
|
|
6
6
|
|
|
7
|
+
## How to use Antigravity Awesome Skills with Claude Code
|
|
8
|
+
|
|
9
|
+
Install the library into Claude Code, then invoke focused skills directly in the conversation or through the plugin marketplace path. Claude Code benefits most when you keep the prompt specific about the skill, the scope, and the intended output.
|
|
10
|
+
|
|
7
11
|
## Why use this repo for Claude Code
|
|
8
12
|
|
|
9
|
-
- It includes 1,
|
|
13
|
+
- It includes 1,328+ skills instead of a narrow single-domain starter pack.
|
|
10
14
|
- It supports the standard `.claude/skills/` path and the Claude Code plugin marketplace flow.
|
|
11
15
|
- It includes onboarding docs, bundles, and workflows so new users do not need to guess where to begin.
|
|
12
16
|
- It covers both everyday engineering tasks and specialized work like security reviews, infrastructure, product planning, and documentation.
|
|
@@ -4,6 +4,14 @@ If you want **Codex CLI skills** that are easy to install and practical in a loc
|
|
|
4
4
|
|
|
5
5
|
Antigravity Awesome Skills supports Codex CLI through the `.codex/skills/` path and gives you a wide set of reusable task playbooks for planning, implementation, debugging, testing, security review, and delivery.
|
|
6
6
|
|
|
7
|
+
## How to use Antigravity Awesome Skills with Codex CLI
|
|
8
|
+
|
|
9
|
+
Install the library into your Codex path, then invoke focused skills directly in your prompt. The most common pattern is:
|
|
10
|
+
|
|
11
|
+
1. install with `npx antigravity-awesome-skills --codex`
|
|
12
|
+
2. choose one workflow-oriented skill such as `@brainstorming`, `@concise-planning`, or `@test-driven-development`
|
|
13
|
+
3. ask Codex to apply that skill to a concrete file, feature, test, or bugfix
|
|
14
|
+
|
|
7
15
|
## Why use this repo for Codex CLI
|
|
8
16
|
|
|
9
17
|
- It supports Codex CLI with a dedicated install flag and a standard skills layout.
|
|
@@ -4,6 +4,10 @@ If you searched for **Cursor skills** on GitHub, this repository is built to be
|
|
|
4
4
|
|
|
5
5
|
Antigravity Awesome Skills supports Cursor through the `.cursor/skills/` path and keeps the entry point simple: install once, then invoke the skills you need in chat.
|
|
6
6
|
|
|
7
|
+
## How to use Antigravity Awesome Skills with Cursor
|
|
8
|
+
|
|
9
|
+
Install the library into Cursor's skills directory, then call skills directly in chat with `@skill-name`. Cursor works especially well when you combine planning, implementation, and validation skills inside one conversation.
|
|
10
|
+
|
|
7
11
|
## Why use this repo for Cursor
|
|
8
12
|
|
|
9
13
|
- It supports Cursor directly with a dedicated install flag.
|
|
@@ -29,6 +29,15 @@ Start from:
|
|
|
29
29
|
- [bundles.md](bundles.md)
|
|
30
30
|
- [workflows.md](workflows.md)
|
|
31
31
|
|
|
32
|
+
### What is the difference between skills and MCP tools?
|
|
33
|
+
|
|
34
|
+
- **Skills** are reusable `SKILL.md` playbooks that guide an AI assistant through a workflow.
|
|
35
|
+
- **MCP tools** are integrations or callable capabilities that let the assistant interact with external systems.
|
|
36
|
+
|
|
37
|
+
Use skills when you want better process, structure, and execution quality. Use MCP tools when you need access to APIs, services, databases, or other systems. Use both when you want reliable workflows plus external capabilities.
|
|
38
|
+
|
|
39
|
+
For the longer explanation, read [skills-vs-mcp-tools.md](skills-vs-mcp-tools.md).
|
|
40
|
+
|
|
32
41
|
### Which AI tools work with these skills?
|
|
33
42
|
|
|
34
43
|
- ✅ **Claude Code** (Anthropic CLI)
|
|
@@ -4,11 +4,15 @@ If you are evaluating **Gemini CLI skills** on GitHub, this repository is a stro
|
|
|
4
4
|
|
|
5
5
|
Antigravity Awesome Skills supports Gemini CLI through the `.gemini/skills/` path and combines general engineering playbooks with specialized skills for AI systems, integrations, infrastructure, testing, product, and growth.
|
|
6
6
|
|
|
7
|
+
## How to use Antigravity Awesome Skills with Gemini CLI
|
|
8
|
+
|
|
9
|
+
Install into the Gemini skills path, then ask Gemini to apply one skill at a time to a specific task. This works best when you keep the active set small and choose a clear workflow-oriented skill for the job in front of you.
|
|
10
|
+
|
|
7
11
|
## Why use this repo for Gemini CLI
|
|
8
12
|
|
|
9
13
|
- It installs directly into the expected Gemini skills path.
|
|
10
14
|
- It includes both core software engineering skills and deeper agent/LLM-oriented skills.
|
|
11
|
-
- It helps new users get started with bundles and workflows rather than forcing a cold start from 1,
|
|
15
|
+
- It helps new users get started with bundles and workflows rather than forcing a cold start from 1,328+ files.
|
|
12
16
|
- It is useful whether you want a broad internal skill library or a single repo to test many workflows quickly.
|
|
13
17
|
|
|
14
18
|
## Install Gemini CLI Skills
|
|
@@ -18,7 +18,7 @@ Kiro is AWS's agentic AI IDE that combines:
|
|
|
18
18
|
|
|
19
19
|
Kiro's agentic capabilities are enhanced by skills that provide:
|
|
20
20
|
|
|
21
|
-
- **Domain expertise** across 1,
|
|
21
|
+
- **Domain expertise** across 1,328+ specialized areas
|
|
22
22
|
- **Best practices** from Anthropic, OpenAI, Google, Microsoft, and AWS
|
|
23
23
|
- **Workflow automation** for common development tasks
|
|
24
24
|
- **AWS-specific patterns** for serverless, infrastructure, and cloud architecture
|
|
@@ -0,0 +1,89 @@
|
|
|
1
|
+
# Skills vs MCP Tools
|
|
2
|
+
|
|
3
|
+
If you are trying to understand the difference between **Antigravity skills** and **MCP tools**, the short version is:
|
|
4
|
+
|
|
5
|
+
- **Skills** are reusable `SKILL.md` playbooks that tell an AI assistant how to execute a workflow.
|
|
6
|
+
- **MCP tools** are integrations or callable capabilities that let the assistant interact with external systems.
|
|
7
|
+
|
|
8
|
+
The two are complementary, not competing.
|
|
9
|
+
|
|
10
|
+
## What a skill does
|
|
11
|
+
|
|
12
|
+
A skill gives the model better instructions for a repeated task such as:
|
|
13
|
+
|
|
14
|
+
- planning a feature
|
|
15
|
+
- reviewing code
|
|
16
|
+
- running a security audit
|
|
17
|
+
- writing a README
|
|
18
|
+
- debugging a failing test suite
|
|
19
|
+
|
|
20
|
+
In practice, a skill improves the assistant's decision-making, structure, and process for a task.
|
|
21
|
+
|
|
22
|
+
Example:
|
|
23
|
+
|
|
24
|
+
- `@brainstorming` helps the model clarify requirements before implementation.
|
|
25
|
+
- `@lint-and-validate` helps the model run the right quality checks before claiming success.
|
|
26
|
+
|
|
27
|
+
## What an MCP tool does
|
|
28
|
+
|
|
29
|
+
An MCP tool gives the model a capability it would not otherwise have, such as:
|
|
30
|
+
|
|
31
|
+
- reading from a database
|
|
32
|
+
- calling GitHub APIs
|
|
33
|
+
- fetching docs from a service
|
|
34
|
+
- creating calendar events
|
|
35
|
+
- querying an external system
|
|
36
|
+
|
|
37
|
+
In practice, an MCP tool expands what the assistant can do in the world.
|
|
38
|
+
|
|
39
|
+
## The easiest mental model
|
|
40
|
+
|
|
41
|
+
Use this rule:
|
|
42
|
+
|
|
43
|
+
- **Skills tell the assistant how to work.**
|
|
44
|
+
- **MCP tools tell the assistant what systems it can touch.**
|
|
45
|
+
|
|
46
|
+
If you only install tools, the assistant may have access but still behave inconsistently.
|
|
47
|
+
|
|
48
|
+
If you only install skills, the assistant may know the workflow but still lack the capability to reach the external system it needs.
|
|
49
|
+
|
|
50
|
+
Together, they are much stronger.
|
|
51
|
+
|
|
52
|
+
## Which one should you start with?
|
|
53
|
+
|
|
54
|
+
Start with **skills** if:
|
|
55
|
+
|
|
56
|
+
- you want better planning, coding, debugging, testing, or review behavior immediately
|
|
57
|
+
- you are working mostly in local files and terminal flows
|
|
58
|
+
- you want reusable playbooks before adding more integrations
|
|
59
|
+
|
|
60
|
+
Start with **MCP tools** if:
|
|
61
|
+
|
|
62
|
+
- your main blocker is access to external systems
|
|
63
|
+
- you need the model to call APIs, query services, or interact with hosted platforms
|
|
64
|
+
- you already like the model's workflow quality, but need more reach
|
|
65
|
+
|
|
66
|
+
Use **both** when:
|
|
67
|
+
|
|
68
|
+
- you want reliable workflows plus external capabilities
|
|
69
|
+
- you are building agent systems, internal tooling, or multi-step operational flows
|
|
70
|
+
|
|
71
|
+
## How this repo fits in
|
|
72
|
+
|
|
73
|
+
Antigravity Awesome Skills is primarily a **skill library**:
|
|
74
|
+
|
|
75
|
+
- installable `SKILL.md` playbooks
|
|
76
|
+
- bundles for role-based starting points
|
|
77
|
+
- workflows for ordered execution patterns
|
|
78
|
+
- tool-specific guides for Claude Code, Cursor, Codex CLI, Gemini CLI, and others
|
|
79
|
+
|
|
80
|
+
Many skills in this repo also explain how to work with MCP, APIs, and other integrations, but the repository itself is centered on reusable workflow guidance rather than acting as an MCP server.
|
|
81
|
+
|
|
82
|
+
## Good next reads
|
|
83
|
+
|
|
84
|
+
- [FAQ](faq.md)
|
|
85
|
+
- [Bundles](bundles.md)
|
|
86
|
+
- [Workflows](workflows.md)
|
|
87
|
+
- [AI Agent Skills](ai-agent-skills.md)
|
|
88
|
+
- [Codex CLI Skills](codex-cli-skills.md)
|
|
89
|
+
- [Gemini CLI Skills](gemini-cli-skills.md)
|
|
@@ -12,7 +12,7 @@ Great question! Here's what just happened and what to do next:
|
|
|
12
12
|
|
|
13
13
|
When you ran `npx antigravity-awesome-skills` or cloned the repository, you:
|
|
14
14
|
|
|
15
|
-
✅ **Downloaded 1,
|
|
15
|
+
✅ **Downloaded 1,328+ skill files** to your computer (default: `~/.gemini/antigravity/skills/`; or a custom path like `~/.agent/skills/` if you used `--path`)
|
|
16
16
|
✅ **Made them available** to your AI assistant
|
|
17
17
|
❌ **Did NOT enable them all automatically** (they're just sitting there, waiting)
|
|
18
18
|
|
|
@@ -32,7 +32,7 @@ Bundles are **recommended lists** of skills grouped by role. They help you decid
|
|
|
32
32
|
|
|
33
33
|
**Analogy:**
|
|
34
34
|
|
|
35
|
-
- You installed a toolbox with 1,
|
|
35
|
+
- You installed a toolbox with 1,328+ tools (✅ done)
|
|
36
36
|
- Bundles are like **labeled organizer trays** saying: "If you're a carpenter, start with these 10 tools"
|
|
37
37
|
- You don't install bundles—you **pick skills from them**
|
|
38
38
|
|
|
@@ -202,7 +202,7 @@ Let's actually use a skill right now. Follow these steps:
|
|
|
202
202
|
|
|
203
203
|
## Step 5: Picking Your First Skills (Practical Advice)
|
|
204
204
|
|
|
205
|
-
Don't try to use all 1,
|
|
205
|
+
Don't try to use all 1,328+ skills at once. Here's a sensible approach:
|
|
206
206
|
|
|
207
207
|
If you want a tool-specific starting point before choosing skills, use:
|
|
208
208
|
|
|
@@ -333,7 +333,7 @@ Usually no, but if your AI doesn't recognize a skill:
|
|
|
333
333
|
|
|
334
334
|
### "Can I load all skills into the model at once?"
|
|
335
335
|
|
|
336
|
-
No. Even though you have 1,
|
|
336
|
+
No. Even though you have 1,328+ skills installed locally, you should **not** concatenate every `SKILL.md` into a single system prompt or context block.
|
|
337
337
|
|
|
338
338
|
The intended pattern is:
|
|
339
339
|
|
|
@@ -34,7 +34,7 @@ antigravity-awesome-skills/
|
|
|
34
34
|
├── 📄 CONTRIBUTING.md ← Contributor workflow
|
|
35
35
|
├── 📄 CATALOG.md ← Full generated catalog
|
|
36
36
|
│
|
|
37
|
-
├── 📁 skills/ ← 1,
|
|
37
|
+
├── 📁 skills/ ← 1,328+ skills live here
|
|
38
38
|
│ │
|
|
39
39
|
│ ├── 📁 brainstorming/
|
|
40
40
|
│ │ └── 📄 SKILL.md ← Skill definition
|
|
@@ -47,7 +47,7 @@ antigravity-awesome-skills/
|
|
|
47
47
|
│ │ └── 📁 2d-games/
|
|
48
48
|
│ │ └── 📄 SKILL.md ← Nested skills also supported
|
|
49
49
|
│ │
|
|
50
|
-
│ └── ... (1,
|
|
50
|
+
│ └── ... (1,328+ total)
|
|
51
51
|
│
|
|
52
52
|
├── 📁 apps/
|
|
53
53
|
│ └── 📁 web-app/ ← Interactive browser
|
|
@@ -100,7 +100,7 @@ antigravity-awesome-skills/
|
|
|
100
100
|
|
|
101
101
|
```
|
|
102
102
|
┌─────────────────────────┐
|
|
103
|
-
│ 1,
|
|
103
|
+
│ 1,328+ SKILLS │
|
|
104
104
|
└────────────┬────────────┘
|
|
105
105
|
│
|
|
106
106
|
┌────────────────────────┼────────────────────────┐
|
|
@@ -201,7 +201,7 @@ If you want a workspace-style manual install instead, cloning into `.agent/skill
|
|
|
201
201
|
│ ├── 📁 brainstorming/ │
|
|
202
202
|
│ ├── 📁 stripe-integration/ │
|
|
203
203
|
│ ├── 📁 react-best-practices/ │
|
|
204
|
-
│ └── ... (1,
|
|
204
|
+
│ └── ... (1,328+ total) │
|
|
205
205
|
└─────────────────────────────────────────┘
|
|
206
206
|
```
|
|
207
207
|
|
package/package.json
CHANGED