Pablan, as it stands

Self-hosted knowledge management for SMEs: a split-screen Markdown editor
whose sections an LLM refines while you write, and RAG question answering
over the documents that result. FastAPI + Postgres/pgvector on the back,
SvelteKit on the front, everything OpenAI-compatible and self-hostable.

Squashed into a single commit; the development history stays local.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CA43ZJda8Rbp2hKXNy8f6b

backend/app/models/prompt_setting.py

@@ -0,0 +1,23 @@
+from sqlalchemy import String, Text
+from sqlalchemy.orm import Mapped, mapped_column
+
+from app.models.base import Base, TimestampMixin, UUIDPrimaryKeyMixin
+
+
+class PromptSetting(UUIDPrimaryKeyMixin, TimestampMixin, Base):
+    """An admin override for a shipped system prompt.
+
+    Every prompt has a CODE default (`app/prompts/defaults.py`); a row here
+    exists only when an admin has changed one. `content` is the full replacement
+    text. Applied without a restart via a module-level cache
+    (`app/prompts/overrides.py`), refreshed on every write — like the LLM
+    settings, and single-process by design (`--workers 1`). Resetting a prompt
+    deletes its row, so the code default takes over again. Unlike LLM settings
+    there is no `.env` layer: prompts have no environment representation, so the
+    reset target is the code default rather than the environment.
+    """
+
+    __tablename__ = "prompt_settings"
+
+    key: Mapped[str] = mapped_column(String(64), unique=True)
+    content: Mapped[str] = mapped_column(Text)