All notable changes to this project will be documented in this file.
- 🛠️ Large runs are now lighter on memory and finish faster, so full-board runs no longer risk running out of memory on the default setting.
- 🔎 Searches and other filters (keywords, locations, technologies, salary, remote, visa, Japanese level) now look across the whole job board instead of only the first few listings, so matching jobs are no longer missed when "Maximum records" is small.
- ⚙️ Renamed the monitoring scope input field to
monitorScope for clarity. Existing scheduled runs that already pass the previous field name continue to work without any changes.
- Source-backed batch queries, locations, technology, salary, remote, visa, and Japanese-level filters.
- Stable job IDs, deterministic sorting, source provenance, explicit failure rows, and named monitor state.
- Snapshot and changes monitoring with
NEW, UPDATED, REAPPEARED, EXPIRED, and UNCHANGED transitions.
- Optional market summary built only from published salary values and currencies.
- Missing source values are omitted instead of defaulting to Tokyo, today's date, Engineering, full-time, JPY, or empty URLs.
- Removed standby server mode and its OpenAPI configuration.
- 🔗 Added quick links to our other 7 Japan jobs actors, so you can find mid-career, graduate, executive, and bilingual tech job leads all in one place.
- ✨ Added Japan-based tech job listings, company profiles, and salary ranges extraction.
- ⚡ Optimized crawler performance and request timeouts.
