| name | Smart Web Clipper |
| description | Evaluates a slash command `/clip <url>`, uses a Python script to fetch the minimized HTML, cleans it of ads/garbage, and saves it as a pristine Markdown note in the vault. |
Smart Web Clipper Workflow
When the user types a slash command like /clip <url>, /save <url>, or asks you to parse a webpage from the internet, follow these steps strictly:
1. Extract the Target URL
Identify the URL provided by the user in their prompt. If no URL is provided, ask the user for one.
2. Run the Fetcher Script
Execute the provided Python helper script via your terminal command tool (run_command) to fetch the raw HTML. The script automatically strips out heavy garbage (scripts, styles, navbars, footers) to save your context window.
python /absolute/path/to/Obsidian-AI-Skills/smart_web_clipper/scripts/fetch_cleaner.py "<URL>"
(Replace /absolute/path/to/ with the actual path to this workspace).
3. Read and Parse the Output
Read the output generated by the script. This output will be minimized HTML.
- Understand the core subject of the webpage (e.g., an article, a recipe, a tutorial).
- Discard any remaining obvious layout boilerplate, cookie banners, or "Read more" widgets that survived the regex scraper.
4. Format into Pristine Markdown
Convert the extracted core content into standard Markdown format:
- Use
#, ## for structural headers.
- Extract
[links](url) and formatting (**bold**, *italic*) correctly based on the HTML <a>, <b>, <i> tags.
- Extract images using
 syntax.
5. Create the Obsidian Note
6. Notify the User
Confirm to the user that the page has been clipped successfully, showing the file name and a 1-sentence summary of what was saved.