working through the plan + UI/ UX

src-tauri/src/llm/mod.rs

@@ -1,6 +1,5 @@
 use serde::{Deserialize, Serialize};
 use std::sync::Mutex;
-use tauri::ipc::Channel;
 
 // ─── LLM Event (for streaming) ───────────────────────────────
 
@@ -16,6 +15,8 @@ pub enum LlmEvent {
 
 pub struct AppState {
     pub config: Mutex<LlmConfig>,
+    pub rag: crate::rag::RagStore,
+    pub gen: crate::generations::GenerationStore,
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
@@ -26,6 +27,10 @@ pub struct LlmConfig {
     pub temperature: f32,
     pub max_tokens: u32,
     pub top_p: f32,
+    /// Ollama image-generation model (e.g. `x/flux2-klein:4b`). macOS-only via Ollama.
+    pub image_model: String,
+    /// Ollama embedding model for lore RAG (e.g. `nomic-embed-text`).
+    pub embed_model: String,
 }
 
 impl Default for LlmConfig {
@@ -38,6 +43,8 @@ impl Default for LlmConfig {
             temperature: 0.7,
             max_tokens: 512,
             top_p: 0.9,
+            image_model: "x/flux2-klein:4b".to_string(),
+            embed_model: "nomic-embed-text".to_string(),
         }
     }
 }
@@ -50,6 +57,10 @@ pub struct GenerateRequest {
     pub system: Option<String>,
     pub temperature: Option<f32>,
     pub max_tokens: Option<u32>,
+    /// Optional RAG query: when set, the top lore chunks for this query are
+    /// retrieved and prepended to the system prompt so generation stays
+    /// consistent with the user's world bible.
+    pub rag_query: Option<String>,
 }
 
 // ─── OpenAI-Compatible Chat Response ─────────────────────────