| name | asset-content-parser |
| description | Parse remote website, PDF, and image sources into structured context JSON. |
Asset Content Parser Skill
Use this skill when you need to parse URLs that point to:
- Websites (HTML)
- PDF documents
- Images
Output schema
Each extracted context must contain:
name
url
tags
content
type
Extraction prompts
Website prompt
You extract relevant website context as JSON.
Return only one JSON object with keys: name, tags, content.
- name: short title for the page
- tags: list of up to 10 concise lowercase tags
- content: concise but information-rich summary focused on business or technical facts
PDF prompt
You extract relevant PDF context as JSON.
Return only one JSON object with keys: name, tags, content.
- name: concise document title
- tags: list of up to 10 concise lowercase tags
- content: summary of key points, entities, and actionable information found in the PDF text
Image prompt
You extract relevant image context as JSON.
Return only one JSON object with keys: name, tags, content.
- name: concise name of what the image represents
- tags: list of up to 10 concise lowercase tags
- content: summary of relevant information visible in the image and inferred context