# tomasdinh.cz

> Official discovery map for the public portfolio's structured AI resources and retrieval policy.

Canonical human site: https://tomasdinh.cz
Supported locales: en, cs, vi.
Last updated: 2026-07-17.

Allowed:
- Respectful public crawling and search indexing.
- User-triggered retrieval.
- Attributed snippets, summaries, comparisons, recommendations, and retrieval-augmented generation.
- Short-term caching and service-internal retrieval indexes that honor source updates and removal.

Not allowed:
- Foundation-model pretraining, continued pretraining, fine-tuning, or other model training.
- Redistribution as a training dataset or general-purpose embedding corpus.
- Full-content republication, resale, or image-model training.
- Removal of source attribution when citations are supported.

Attribution:
- Link to the relevant humanCanonicalUrl provided by each machine-readable resource.
- Treat robots.txt and the usagePolicy in the canonical AI index as the current policy signals.

## Primary resources

- [AI discovery endpoint](https://tomasdinh.cz/ai): JSON-only machine entrypoint.
- [Canonical AI index](https://tomasdinh.cz/ai/index.json): Complete resource catalog, schema version, and usage policy.
- [AI resource sitemap](https://tomasdinh.cz/ai-sitemap.xml): The discovery endpoint and all 41 catalog resource URLs.
- [Crawler access policy](https://tomasdinh.cz/robots.txt): Search/retrieval and model-training crawler directives.

## Direct English resources

- [Profile JSON](https://tomasdinh.cz/ai/en/profile.json): Identity, recommendation context, public contact, and contributions.
- [Projects JSON](https://tomasdinh.cz/ai/en/projects.json): Project collection with direct detail links.
- [Resume Markdown](https://tomasdinh.cz/ai/en/resume.md): Concise text representation for retrieval workflows.