So I created a web clipper template for GitHub.
Here is my full template JSON:
{
"schemaVersion": "0.1.0",
"name": "GitHub Repository",
"behavior": "create",
"noteContentFormat": "{{\"Insert a 'Repository Overview' H2 markdown heading that contains a general summary of the repository, including a description and overview of the repository's purpose and what it's used for\"}}\n\n{{\"Insert a 'Installation and Configuration' H2 markdown heading that contains specific steps on how to install, setup, configure, and use the tool or service if available\"}}\n\n{{\"Insert a 'Examples' H2 markdown heading that contains specific code examples on how to use the tool or service described in the repository if the repository contains specific example code. Provide 3-10 code examples of how to install or set up the tool/service if available.\"}}\n\n## README\n\n{{selectorHtml:article|markdown}}\n",
"properties": [
{
"name": "title",
"value": "{{selector:#repo-title-component a}}",
"type": "text"
},
{
"name": "owner",
"value": "{{meta:name:octolytics-dimension-user_login}}",
"type": "text"
},
{
"name": "source",
"value": "{{url}}",
"type": "text"
},
{
"name": "website",
"value": "{{selector:.Layout-sidebar a?href|first}}",
"type": "text"
},
{
"name": "description",
"value": "{{selector:.Layout-sidebar p|first}}",
"type": "text"
},
{
"name": "stars",
"value": "{{selector:#repo-stars-counter-star}}",
"type": "number"
},
{
"name": "tags",
"value": "github, repo, {{selector:.topic-tag|join:\\\", \\\"}}",
"type": "multitext"
},
{
"name": "created",
"value": "{{date}}",
"type": "date"
}
],
"triggers": [
"https://github.com/"
],
"noteNameFormat": "{{meta:name:octolytics-dimension-user_login}} {{selector:#repo-title-component a}}: {{selector:.Layout-sidebar p|first}}",
"path": "00 Clipped",
"context": "{{content|markdown}}"
}
When I run the clipper on a random repo (chasebank87/mesh-ai) for example, I get the following markdown:
---
title: "mesh-ai"
owner: "chasebank87"
source: "https://github.com/chasebank87/mesh-ai"
website: "#readme-ov-file"
description:
stars: 141
tags:
- "github"
- "repo"
created: 2026-02-14
---
## Repository Overview
Mesh AI is an Obsidian plugin that brings the functionality of the [Fabric tool](https://github.com/danielmiessler/fabric) directly into your Obsidian vault. It enables users to run and manage multiple AI-powered text processing patterns on their notes, clipboard content, or Tavily search results.
The plugin integrates with major AI service providers including OpenAI, Google, Microsoft, Anthropic, Groq, and Ollama to enhance productivity within Obsidian. Key capabilities include sequential pattern processing (chaining outputs), pattern stitch mode (collective results), automatic YouTube transcript fetching, custom workflow creation, and pattern management with the ability to download patterns from the Fabric repository or maintain custom patterns.
This is a sibling project to the Unofficial Fabric Plugin, with the key advantage that it requires no additional dependencies—Fabric or Fabric Connector installation is not needed.
## Installation and Configuration
### Installation Methods
**Method 1: Manual Installation**
1. Download the latest release from the [GitHub repository](https://github.com/chasebank87/mesh-ai/releases)
2. Extract the files into your Obsidian vault's `plugins` directory (`.obsidian/plugins/mesh-ai/`)
3. Enable the plugin from Obsidian Settings → Community Plugins
**Method 2: BRAT Plugin Installation**
1. Install the BRAT Plugin from the Obsidian Community Plugins list
2. Open BRAT settings and add the repository URL: `https://github.com/chasebank87/mesh-ai`
3. Follow the prompts to install and enable the Mesh AI plugin
### Configuration
1. Open Obsidian Settings → Mesh AI
2. Configure your preferred AI service provider (OpenAI, Groq, Ollama, etc.)
3. Enter your API credentials for the selected provider
4. Download patterns from the Fabric repository or set up a custom pattern folder
5. Define text processing patterns and workflows
### Usage
1. Use the command palette to apply workflows to notes, clipboard, or Tavily search results
2. Create custom workflows from the settings view for quick command palette access
3. Access the Mesh AI Interface for interactive processing
## Examples
The repository does not contain specific code examples for end-user implementation, as this is an Obsidian plugin that is configured through the UI rather than programmatically. However, here are the key configuration and usage patterns:
```bash
# Clone the repository for development
git clone https://github.com/chasebank87/mesh-ai.git
cd mesh-ai
```
```bash
# Install dependencies (for development)
npm install
```
```bash
# Build the plugin (for development)
npm run build
```
```
# Manual installation path structure
.obsidian/
plugins/
mesh-ai/
main.js
manifest.json
styles.css
```
```
# BRAT repository URL for installation
https://github.com/chasebank87/mesh-ai
```
Note: Since Mesh AI is a GUI-based Obsidian plugin, most configuration is done through the Obsidian settings interface rather than through code. The plugin handles AI provider integration, pattern management, and workflow execution internally.
## README
# Table of Contents- [Mesh AI](https://github.com/chasebank87/#mesh-ai)
- [Table of Contents](https://github.com/chasebank87/#table-of-contents)
- [Overview](https://github.com/chasebank87/#overview)
- [Features](https://github.com/chasebank87/#features)
- [Compatibility](https://github.com/chasebank87/#compatibility)
- [Under Development](https://github.com/chasebank87/#under-development)
- [Differences from Official Fabric Plugin](https://github.com/chasebank87/#differences-from-official-fabric-plugin)
- [Installation](https://github.com/chasebank87/#installation)
- [Usage](https://github.com/chasebank87/#usage)
- [Contributing](https://github.com/chasebank87/#contributing)
- [License](https://github.com/chasebank87/#license)
## OverviewThe Mesh AI plugin emulates the functionality provided by the [Fabric tool](https://github.com/danielmiessler/fabric) within your Obsidian vault. This plugin allows users to run and manage multiple text processing patterns on their notes, the clipboard, or Tavily search results. It integrates with various AI services and platforms to enhance your productivity within Obsidian.
## Features- **Process Active Note, Clipboard, or Tavily Search Results:** The plugin allows you to apply patterns to the currently active note, clipboard content, or results from Tavily search.
- **Pattern Processing:**
- **Sequential Mode (Default):** Process multiple patterns in order, where the output of the first pattern is used as the input for the next.
- **Pattern Stitch Mode:** Process multiple patterns and output the results of each pattern collectively.
- **YouTube Link Detection:** Automatically detects YouTube links, fetches their transcripts, and uses the transcripts as a source input for processing.
- **Workflow Creation:** Create workflows from the settings view, which can then be added as command palette items for quick access.
- **Pattern Management:**
- Download patterns from the Fabric repository.
- Maintain a custom pattern folder to manage your own patterns.
- **AI Service Integration:** Utilize various major providers such as:
- `openai`
- `google`
- `microsoft`
- `anthropic`
- `grocq`
- `ollama`
## CompatibilityCurrently, the plugin has been tested and confirmed to work with:
- `Grocq`
- `OpenAI`
- `Ollama`
### Under DevelopmentWe are currently working on testing and development for:
- `Microsoft Azure AI Services`
- `Anthropic`
- `google`
## Differences from Official Fabric PluginThis plugin is a sibling project to the [Unofficial Fabric Plugin](https://github.com/chasebank87/unofficial-fabric-plugin). The key differences include:
- **Visual Appearance:** Differences in the user interface.
- **No Additional Dependencies:** You do not need to have Fabric or Fabric Connector installed for this plugin to work.
## InstallationTo install the plugin, follow these steps:
1. Download the latest release from the [GitHub repository](https://github.com/chasebank87/unofficial-fabric-plugin).
2. Extract the files into your Obsidian vault's `plugins` directory.
3. Enable the plugin from the Obsidian settings menu.
\-OR-
Install using the BRAT Plugin:
1. Install the BRAT Plugin from the Obsidian Community Plugins list.
2. Open the BRAT settings and add the repository URL: `https://github.com/chasebank87/meshl-ai`.
3. Follow the prompts to install and enable the Mesh AI plugin.
## Usage1. Open the settings view within Obsidian and configure your preferred AI service provider.
2. Define your text processing patterns and workflows.
3. Use the command palette to apply your workflows to your notes, clipboard, or Tavily search results.
4. Use the Mesh AI Interface
## ContributingContributions are welcome! Please feel free to submit a pull request or open an issue if you encounter any problems or have suggestions for new features.
## LicenseThis project is licensed under the MIT License. See the [LICENSE](https://github.com/chasebank87/mesh-ai/blob/main/LICENSE) file for details.
---
For more information, please visit our [GitHub repository](https://github.com/chasebank87/mesh-ai).
---
Notice how the headers run together:
## OverviewThe Mesh AI plugin emulates the functionality provided by the [Fabric tool](https://github.com/danielmiessler/fabric) within your Obsidian vault. This plugin allows users to run and manage multiple text processing patterns on their notes, the clipboard, or Tavily search results. It integrates with various AI services and platforms to enhance your productivity within Obsidian.
Image:
This is how it should be formatted:
What’s wrong with my template? How can I get the README’s headers and content to render correctly?
Is my Interpreter Context incorrect? What am I doing incorrectly?