⚡ Auto-Pilot Scribing
You don't always need to type a prompt. Scribe & Saga supports dynamic auto-advancements:
- Empty Input + Blank Page: Pressing Send (or Ctrl+Enter) with zero history triggers a silent, initial
"start" message to establish your story.
- Empty Input + Ongoing Story: Submitting an empty field during a chronicle sends a quiet background
"continue" command, prompting the AI storyteller to draft the next sequence seamlessly.
🎭 Classic Novel Formatting
The engine parses LLM outputs in real-time, giving standard manuscript rules beautiful typographic distinction:
"I advise we move quickly,"
-> Quotes render in warm amber italics.
*The floorboards groan heavily beneath your boots*
-> Asterisk prose converts to muted action font.
💭 Reasoning Analysis: Evaluating physical environment...
-> <think> blocks are structured into reasoning modules (great for DeepSeek R1 models).
⏳ Editing & Branching Timelines
Hovering over any manuscript block reveals chronological modification tools:
- ✍️ Rewrite: Copies your corresponding prompt back into the action bar and slices off everything beyond that point, letting you redirect the timeline.
- ✂️ Prune: Instantly severs the current narrative sequence from that specific block forward, establishing a clean slate from your last favored event.
📡 Local Connection Setup (CORS Override)
Web browsers isolate script requests from online sandboxes to local hostnames. To authorize the engine to speak with your local Ollama port, launch Ollama with a cross-origin directive:
On Windows (Command Prompt):
set OLLAMA_ORIGINS="*"
ollama serve
On macOS (Terminal launch for desktop app):
OLLAMA_ORIGINS="*" open -a Ollama
(Quit the running menu-bar app first before running this terminal launcher!)
On Linux (Systemd Service):
sudo systemctl edit ollama.service
Add these rows under the Service section, save, and restart:
[Service]
Environment="OLLAMA_ORIGINS=*"