How we built a travel guide from open data with Claude Code

Описание и применение

Wanloria – это бесплатное приложение для путешествий, которое использует проверенные данные из OpenStreetMap, Wikidata и официальных сайтов, генерирует описания только из проверенных полей, рассчитывает время походов по DIN 33466, фильтрует треки по строгим критериям и публикует сервер‑рендеренные страницы с структурированными данными.

Задача

Travel content is mostly unverifiable; most pages copy from other pages without source traceability.

Результат

Wanloria delivers travel content that is fully verifiable, with computed hiking times, difficulty ratings, and server‑rendered pages for SEO, while maintaining a lightweight SPA experience.

Рабочий процесс

• 1. Collect verified data from OpenStreetMap, Wikidata/Wikipedia, official websites, and Open‑Meteo elevation.
• 2. Store each fact with its source URL.
• 3. Use Rule 1: generate prose only from verified fields via an LLM, never copy source sentences.
• 4. Use Rule 2: compute hiking times with DIN 33466 formula and difficulty with MIDE scale; filter trails by signposting, distance, day‑length, and geometry match.
• 5. Use Rule 3: server‑render every public page with JSON‑LD, hreflang, and sitemap.
• 6. Publish SPA with SPA‑backed pages and provide user feedback channel for corrections.

Доказательства

Most "things to do in X" pages are rewritten from other "things to do in X" pages. We wanted every fact on a page (opening hours, price, website, whether a trail is circular) to trace back to a source, or not be shown at all.

Источник

Rule 1: the LLM writes prose, not facts. Descriptions are generated by an LLM only from the verified fields of that place, never copying sentences from the source.

Источник

Rule 2: computed numbers use published formulas. Hiking time follows DIN 33466…

Источник

Rule 3: pages for humans and for crawlers. The app itself is a SPA, but every public page is server‑rendered HTML with JSON‑LD…

Источник