Menü
How to discover and fetch full markdown content of articles published on bnsaed.com.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
How an AI agent should contact Abdelrahman Saed, or book a meeting with him, on behalf of a user.
Overview of Abdelrahman Saed's portfolio site (bnsaed.com) — who he is, what's on the site, and how agents should retrieve structured data.
| name | article-reader |
| description | How to discover and fetch full markdown content of articles published on bnsaed.com. |
| version | 1.0.0 |
| author | Abdelrahman Saed |
| homepage | https://bnsaed.com/#articles |
Help agents find and read the full text of technical articles published by Abdelrahman Saed without scraping the single-page-app HTML.
The canonical list of articles is in the sitemap:
GET https://bnsaed.com/sitemap.xml
Any <loc> under /articles/ is an article page. The slug is the last path segment.
Each article is pre-rendered as a standalone HTML page at /articles/:slug/. Fetch it with the markdown Accept header to get clean markdown:
GET https://bnsaed.com/articles/{slug}
Accept: text/markdown
The response is text/markdown with the article body. An x-markdown-tokens header, if present, estimates token count.
Without the Accept: text/markdown header, you receive the normal HTML page. The HTML includes a <script type="application/ld+json"> block with schema.org/BlogPosting metadata — headline, description, datePublished, author, keywords — which is the fastest way to get structured metadata without parsing the body.
| Slug | Title |
|---|---|
flutter-cached-pdfview-guide | Building a Cached PDF Viewer in Flutter |
Always re-check /sitemap.xml for the current list — new articles are added without updating this skill.
Per the site's Content-Signal in robots.txt: this content is opted in for ai-input=yes (use in a user-facing answer) but opted out of ai-train=no (do not use to train a model). Attribute the author when quoting.