Developer resources

If you are writing a crawler, agent, or small tool that needs to understand nhanaz.io.vn, start with llms.txt. The longer context is in llms-full.txt, machine-readable identity is in entity.json, and canonical public URLs are in sitemap.xml.

The “when to use this site” guide is in agent-instructions.md, with a short alias at agent.md. The Agent Skills index at /.well-known/agent-skills/index.json only describes read access to the archive. The ARD catalog at /.well-known/ai-catalog.json also points only to Markdown documentation, not to tools with write access.

This site exposes read-only MCP, A2A, and NLWeb endpoints, all limited to public archive content. It still has no user account, authentication protocol, GraphQL, webhook, or endpoint that changes data. The GET endpoint /api/v1/views only increments an aggregate view counter, accepts no token, and cannot edit content. Do not send credentials or private data to its URLs. The auth.md, pricing.md, and API catalog notes make that scope explicit. For code and change history, use the GitHub repository. For the writer, use the contact page.

The short Markdown entry point is index.md. Public URLs are served as static HTML, text, or JSON, and the main content does not require JavaScript. Giscus comments are the exception because they load from an external service when a reader opens them.