Skill Porter converts Claude Code skills to Gemini CLI extensions and vice versa. Developers benefit by writing once and deploying to both platforms. It connects to Claude Code and Gemini CLI workflows.
git clone https://github.com/jduncan-rva/skill-porter.gitThe skill-porter is a universal automation tool designed to seamlessly convert Claude Code skills into Gemini CLI extensions and vice versa. This functionality allows developers and AI practitioners to enhance their workflows by integrating different coding environments, facilitating a smoother transition between platforms. By utilizing this skill, users can leverage the power of both Claude Code and Gemini CLI, maximizing their productivity and efficiency in automation tasks. One of the key benefits of the skill-porter is its ability to save time during the development process. While the exact time savings are currently unknown, the tool's streamlined conversion process significantly reduces the manual effort required to adapt skills across different systems. This is particularly advantageous for teams looking to implement AI automation in their projects, as it minimizes the friction associated with switching between different coding frameworks. This skill is particularly suited for developers, product managers, and AI practitioners who are involved in workflow automation. By integrating skill-porter into their toolkit, these professionals can enhance their existing capabilities and improve their project outcomes. For instance, a developer working on a data engineering project can quickly adapt existing Claude Code skills for use in a Gemini CLI environment, thereby accelerating the development cycle and reducing time to market. Implementing the skill-porter is classified as an intermediate task, taking approximately 30 minutes to set up. Users should have a basic understanding of both Claude Code and Gemini CLI to fully leverage the tool's capabilities. As organizations increasingly adopt AI-first workflows, the skill-porter plays a crucial role in bridging the gap between different automation platforms, ensuring that teams can work more efficiently and effectively in their AI initiatives.
[{"step":"Prepare your source skill. Gather the Claude Code skill files (e.g., skills/[SKILL_NAME]/*.py, skills/[SKILL_NAME]/requirements.txt) and ensure it runs successfully in your local environment.","tip":"Test the skill in Claude Code first to verify its functionality. Use `skill run [SKILL_NAME]` to confirm it works as expected."},{"step":"Run the Skill Porter conversion. Use the prompt template to convert the skill to a Gemini CLI extension. Replace [CLAUDE_CODE_SKILL] with the actual skill code or a file path to the skill directory.","tip":"For large skills, break the code into chunks and convert them separately to avoid errors. Use the `--output-dir` flag in the Gemini CLI to specify a directory for the converted files."},{"step":"Review and test the converted extension. Check the generated manifest.json and index.js for accuracy. Ensure all dependencies are listed in the requirements.txt or package.json.","tip":"Run `gemini extension test` to validate the extension. Use `gemini extension install` to load it into your local environment for testing."},{"step":"Deploy the extension. Push the converted files to a Git repository or package them for distribution. Share the extension with your team or publish it to the Gemini CLI extension registry.","tip":"Use `gemini extension publish` to share the extension publicly. Include a README.md with installation instructions and examples."},{"step":"Maintain cross-platform compatibility. Update the skill in one platform and use Skill Porter to regenerate the extension for the other platform whenever changes are made.","tip":"Set up a CI/CD pipeline to automate the conversion process. Use tools like GitHub Actions to run Skill Porter on every commit and deploy updates automatically."}]
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/jduncan-rva/skill-porterCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Convert the following [CLAUDE_CODE_SKILL] into a [GEMINI_CLI_EXTENSION]. Ensure the skill's functionality remains identical, including dependencies, commands, and input/output behavior. Preserve the original [SKILL_NAME] and [DESCRIPTION]. Output the converted extension in a ready-to-use format for the Gemini CLI, including the required manifest.json and any necessary configuration files. [CLAUDE_CODE_SKILL]: ```[SKILL_CODE]```
```json
{
"manifest.json": {
"name": "document-summarizer",
"description": "Summarizes uploaded documents into concise bullet points",
"version": "1.0.0",
"commands": [
{
"name": "summarize",
"description": "Summarize a document",
"parameters": [
{
"name": "file",
"type": "string",
"description": "Path to the document file"
},
{
"name": "length",
"type": "string",
"enum": ["short", "medium", "long"],
"default": "medium"
}
]
}
]
},
"index.js": "const fs = require('fs');
const { summarizeText } = require('./summarizer');
module.exports = {
commands: [
{
name: 'summarize',
description: 'Summarize a document',
run: async ({ file, length }) => {
const content = fs.readFileSync(file, 'utf8');
const summary = summarizeText(content, length);
console.log(summary);
}
}
]
};
// Example output:
// Document: /reports/2023-q4.pdf
// Length: medium
// Summary:
// - Revenue grew 15% YoY to $2.3M
// - Customer acquisition cost decreased by 8%
// - Top market: North America (45% of revenue)
```Simplify cloud deployment
Create Beautiful Branded Deliverables. Ship Professionally. Repeat.
Google's multimodal AI model and assistant
AI assistant built for thoughtful, nuanced conversation
Enterprise workflow automation and service management platform
Automate your spreadsheet tasks with AI power
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan