genexus-skills

Repository with AI skills and references for GeneXus-oriented development workflows, providing compact, actionable, and reusable guidance for coding agents.

---

STRUCTURE

Root
├─ nexa/                     : Nexa skill
│  ├─ references/
│  │  ├─ common-*.md         : shared components over other references
│  │  ├─ global-*.md         : shared global definitions
│  │  └─ object-*.md         : object-specific specifications
│  └─ SKILL.md               : skill definition and workflow
│
├─ frontend/                 : Frontend skills group
│  ├─ chameleon-controls-library/
│  │  ├─ references/         : 58 web component docs
│  │  └─ SKILL.md
│  ├─ design-system-builder/
│  │  ├─ references/         : DS architecture guides
│  │  └─ SKILL.md
│  ├─ mercury-design-system/
│  │  ├─ references/         : Mercury tokens, bundles, icons
│  │  └─ SKILL.md
│  ├─ ui-creator/
│  │  ├─ references/         : UI generation templates
│  │  └─ SKILL.md
│  └─ README.md              : group description
│
├─ other-skills/             : additional skill groups
│  ├─ skill-a/               : individual skill
│  │  ├─ references/         : references for that skill
│  │  ├─ scripts/            : optional helpers
│  │  ├─ assets/             : optional templates/assets
│  │  └─ SKILL.md            : specific skill definition
│  ├─ skill-b/
│  │  └─ ...
│  └─ README.md              : group description
│
├─ .gitignore                : repository ignore patterns
└─ README.md                 : repository entry point

Repository design principles:

• Keep each skill self-contained: SKILL.md + optional references/, scripts/, and assets/.
• Use consistent naming to reduce ambiguity and retrieval cost.
• Prefer progressive disclosure; keep SKILL.md short and define details in references/
• Avoid duplicate guidance across files; keep one source of truth per topic.

---

REFERENCES

All skill-related contributions must follow these baseline references:

• Anthropic Claude Code: Overview
• Anthropic Claude Code: Agent Skills
• OpenAI Codex: Overview
• OpenAI Codex CLI: Skills Guide
• OpenCode: Overview
• OpenCode: Skills

---

ONBOARDING & INSTALLATION

Each platform handles skills with small differences. This repository follows a portable pattern, but always confirm the latest setup in the provider docs listed in REFERENCES.

Step 1: Clone once in any local directory

• Choose a local path to clone the repo; for example:
  • macOS/Linux: ~/workspace/genexus-skills
  • Windows: %USERPROFILE%\workspace\genexus-skills
• Clone repository:
  • macOS/Linux

    git clone https://github.com/genexuslabs/genexus-skills.git ~/workspace/genexus-skills
    

  • Windows

    git clone https://github.com/genexuslabs/genexus-skills.git %USERPROFILE%\workspace\genexus-skills
    

Step 2: Create symlinks from provider folders to desired skills

• First, identify provider skill paths:
  • Claude Code
    • Windows: %USERPROFILE%\.claude\skills
    • macOS/Linux: ~/.claude/skills
  • Codex (OpenAI)
    • Windows: %USERPROFILE%\.codex\skills
    • macOS/Linux: ~/.codex/skills
  • CODA-CLI
    • Windows: %USERPROFILE%\.coda\skills
    • macOS/Linux: ~/.coda/skills
    • Note: Require version 1.5.0 or higher
  • OpenCode
    • Windows: %USERPROFILE%\.config\opencode\skills
    • macOS/Linux: ~/.config/opencode/skills
• Then, create one symlink per skill:
  • macOS/Linux example:

    ln -s ~/workspace/genexus-skills/nexa ~/.coda/skills
    

  • Windows example:

    mklink /J %USERPROFILE%\.coda\skills %USERPROFILE%\workspace\genexus-skills\nexa
    

  • ⚠️ Update both the provider skill path and the skill folder name in each command as needed

Step 3: Validation

• Validate that each provider detects linked skills
  • Claude Code
    • Run claude list-skills
  • Codex (OpenAI)
    • Open interactive session and run /skills
  • CODA-CLI (beta)
    • Open interactive session and run /skills, and then select List available skills
  • OpenCode
    • Open interactive session and verify skills in `` from the native skill tool description
• Confirm skill folder name from Step 1 is displayed the output

Step 4 (optional): OpenCode + GEAI

Integrate OpenCode with Globant Enterprise AI (GEAI)

• Configuration file:
  • Windows: %USERPROFILE%\.config\opencode\opencode.json
  • macOS/Linux: ~/.config/opencode/opencode.json
  • Content:

    {
    	"$schema": "https://opencode.ai/config.json",
    	"provider": {
    		"geia": {
    			"npm": "@ai-sdk/openai-compatible",
    			"name": "Globant Enterprise AI",
    			"options": {
    				"baseURL": "https://api.saia.ai"
    			},
    			"models": {
    				"anthropic/claude-opus-4-6": {
    					"name": "GEIA Claude Opus 4.6"
    				},
    				"anthropic/claude-sonnet-4-6": {
    					"name": "GEIA Claude Sonnet 4.6"
    				},
    				"anthropic/claude-haiku-4-5": {
    					"name": "GEIA Claude Haiku 4.5"
    				},
    				"openai/gpt-5.2-codex": {
    					"name": "GEIA GPT 5.2 Codex"
    				},
    				"xai/grok-code-fast-1": {
    					"name": "GEIA Grok Code Fast"
    				}
    			}
    		}
    	},
    	"model": "anthropic/claude-haiku-4-5"
    }
    

• Credentials file:
  • Windows: %USERPROFILE%\.local\share\opencode\auth.json
  • macOS/Linux: ~/.local/share/opencode/auth.json
  • Content:

    {
    	"geia": {
    		"type": "api",
    		"key": ""
    	}
    }
    

Step 5: Run the skill from any CLI provider

After linking and validating the skill, open the coding CLI you set up and start a session in your project folder.

Once you open your CLI, start typing your skill-related request:

╔═════════════════════════════════════════╗ ║ ○ ○ ║ ║ YOUR CODING-CLI INTERFACE ║ ║ ○ ○ ║ ╚═════════════════════════════════════════╝

%type-your-request-here%

Examples that trigger nexa skill:

• Create a Transaction object for Product with ProductId, ProductName, ProductPrice
• Review the AddCart procedure and apply GeneXus best practices
• Generate a Data Provider from Customer but based in SDT and add output example

---

CONTRIBUTION

Refer to CONTRIBUTING.

---

EXTENSIBILITY

This project is intentionally extensible through Forks.

Fork policy:

• You may fork and extend the repository freely.
• Fork owners are responsible for keeping their fork aligned with upstream updates.
• Upstream sync should be frequent to reduce merge complexity and drift.

Extensibility best practices:

• Prefer additive changes (new sibling references, new skill folders) over invasive rewrites.
• Extend existing sections only when semantics are truly shared.
• Isolate variant-specific guidance in separate files to minimize merge conflicts.
• Preserve compatibility in naming and structure so changes remain easy to rebase.

---

LICENCE

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

   http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.