Move the polish instructions into their own module so the prompt the app sends, the unit tests and the local-model probe use the same text. Tests now pin the two rules that matter: fix a misheard name or in-house term to the learned spelling, and never swap an ordinary word on similarity alone. Verified against the local model with the real learned corpus: 하람록에 was rewritten to HaramLog에 and the dictionary pronunciation 디에로 to D3RO, while a natural sentence with no learned term came back with no word substitutions. Recorded the trap that this model returns empty content unless thinking is disabled. |
||
|---|---|---|
| .. | ||
| 00-index.md | ||
| 01-system-overview.md | ||
| 02-infrastructure.md | ||
| 03-shared-packages.md | ||
| 04-desktop-app.md | ||
| 05-web-app.md | ||
| 06-mobile-app.md | ||
| 07-api-server.md | ||
| 08-admin-console.md | ||
| 09-supabase-backend.md | ||
| 10-feature-catalog.md | ||
| 11-gap-backlog.md | ||
| 12-update-protocol.md | ||