Fix SSE route dependency and align architecture docs

backend/app/api/models/agent.py

@@ -0,0 +1,79 @@
+"""Define API models for agent endpoints."""
+
+from __future__ import annotations
+
+from typing import Dict, List, Optional
+
+from pydantic import BaseModel, Field
+
+# Group agent transport schemas together so route modules stay focused on HTTP flow.
+
+
+class AskRequest(BaseModel):
+    """Define the Ask Request API model."""
+
+    query: str = Field(..., min_length=1, max_length=2000)
+    filters: Optional[str] = None
+    provider: Optional[str] = None
+    model: Optional[str] = None
+    top_k: Optional[int] = Field(default=None, ge=1, le=20)
+    prompt_template: Optional[str] = None
+
+
+class AskResponse(BaseModel):
+    """Define the Ask Response API model."""
+
+    answer: str
+    sources: List[Dict] = Field(default_factory=list)
+    model: str = ""
+    latency_ms: int = 0
+    retrieved_count: int = 0
+    context_tokens: int = 0
+    truncated: bool = False
+    error: Optional[str] = None
+
+
+class ChatRequest(BaseModel):
+    """Define the Chat Request API model."""
+
+    query: str = Field(..., min_length=1, max_length=2000)
+    session_id: Optional[str] = None
+    filters: Optional[str] = None
+    provider: Optional[str] = None
+    model: Optional[str] = None
+    top_k: Optional[int] = Field(default=None, ge=1, le=20)
+
+
+class ChatResponse(BaseModel):
+    """Define the Chat Response API model."""
+
+    session_id: str
+    answer: str
+    sources: List[Dict] = Field(default_factory=list)
+    model: str = ""
+    latency_ms: int = 0
+    message_count: int = 0
+
+
+class SessionInfo(BaseModel):
+    """Define the Session Info API model."""
+
+    session_id: str
+    message_count: int
+    created_at: int
+    updated_at: int
+
+
+class FeedbackRequest(BaseModel):
+    """Define the Feedback Request API model."""
+
+    session_id: str
+    message_index: int
+    rating: int = Field(..., ge=1, le=5)
+    comment: Optional[str] = None
+
+
+class TemplateListResponse(BaseModel):
+    """Define the Template List Response API model."""
+
+    templates: Dict[str, str]