feat(db): Add Postgres db through Neon

refactor: Update the prompt to use XML tags and update the instructions
fix(products): Update fetch_products tool to return a proper json

main.py

@@ -1,101 +1,47 @@
 from agno.agent import Agent
 from agno.models.google import Gemini
-from agno.db.sqlite import SqliteDb
-from agno.tools.reasoning import ReasoningTools
-from agno.tools import tool
+from agno.db.postgres import PostgresDb
+from agno.os import AgentOS
+from agno.memory import MemoryManager
 
-from retriever import PineconeHybridRetriever, FilterModel, ProductItem
+from prompts import SHOPPING_AGENT_PROMPT
+from tools import fetch_products
 
-from typing import Optional
 from dotenv import load_dotenv
+import os
 load_dotenv()
 
+postgres_db = PostgresDb(
+    db_url=os.getenv("NEON_KEY"),
+    memory_table="user_memories",
+)
 
-@tool
-def fetch_products(
-    query: str, limit: int = 5, filters: Optional[FilterModel] = None
-) -> list[ProductItem]:
-    """
-    Fetch products from database based on query and filters
-    
-    Args:
-        query (str): Search query
-        limit (int, optional): Number of products to return. Defaults to 5.
-        filters (FilterModel, optional): Filters to apply. Defaults to None.
-    
-    Returns:
-        list[ProductItem]: List of products matching the query and filters
-    """
-    retriever = PineconeHybridRetriever("amazon-products-catalog")
-    return retriever.search_products(
-        query=query,
-        filters=filters,
-        limit=limit,
-        use_hybrid_search=True,
-        enable_reranking=True,
-    )
-
-SYSTEM_PROMPT = """
-# ROLE
-
-You are a personalized shopping AI Agent. Your goal is to help user find the best products based on their needs.
-
-## Instructions
-
-1.  **Analyze the User Need**: Chat with the user to help them identify the product, along with any specific constraints like price, rating, or category.
-2.  **Fetch Products**: Once they have given you the details about the product they want, use the `fetch_products` tool to get the products.
-        * **Extract `filters`**: Identify any explicit constraints and populate the `filters` object.
-            -   **Price**: Look for terms like "under $50", "over 100 dollars", "between $20 and $60".
-            -   **Rating**: Look for terms like "highly rated", "4 stars or more", "best reviewed". A generic term like "best" can imply a `min_rating` of `4.0`.
-            -   **Rating Count**: Look for terms like "popular", "most reviewed". A generic term like "popular" can imply a `min_reviews` of `100`.
-            -   **Categories**: Identify the categories from the user query. **MUST** be from the allowed categories listed below. Do not invent new categories. For example, if the user says "I need black shirts for men", then the right category would be "men's clothing".
-3.  **Present Products**: Analyze the retrieved products, and present them to user in a nice human-readable format. Provide recommendations to the user.
-
-## Allowed Categories
-
-You must use one of the following category values if a category filter is applicable:
-- appliances
-- car & motorbike
-- tv, audio & cameras
-- sports & fitness
-- grocery & gourmet foods
-- home & kitchen
-- pet supplies
-- stores
-- toys & baby products
-- kids' fashion
-- bags & luggage
-- accessories
-- women's shoes
-- beauty & health
-- men's shoes
-- women's clothing
-- industrial supplies
-- men's clothing
-- music
-- home, kitchen, pets
-
-NOTE: These categories are case-sensitive. So you must use the exact category values provided.
-"""
-
-db = SqliteDb(db_file="tmp/data.db", memory_table="user_memories")
-tools = [ReasoningTools(add_instructions=True), fetch_products]
+tools = [fetch_products]
 
-agent = Agent(
-    model=Gemini(id="gemini-2.0-flash"),
-    markdown=True,
-    db=db,
+shopping_agent = Agent(
+    model=Gemini(id="gemini-2.5-flash"),
+    name="Gem - Shopping Agent",
+    id="shopping-agent",
+    add_datetime_to_context=True,
+    db=postgres_db,
     tools=tools,
-    instructions=SYSTEM_PROMPT,
+    instructions=SHOPPING_AGENT_PROMPT,
     enable_agentic_memory=True,
+    memory_manager=MemoryManager(
+        Gemini(id="gemini-2.5-flash-lite"),
+        db=db,
+    ),
     add_history_to_context=True,
     num_history_runs=10,
 )
 
-agent.cli_app(
-    input="Hi. I am looking for something to wear in winter.",
-    session_id="session-01",
-    user_id="user-01",
-    stream=True,
+agent_os = AgentOS(
+    os_id="shopping-agent-os",
+    description="A shopping Agent OS that helps users find the best products based on their needs.",
+    agents=[shopping_agent]
+)
+
+app = agent_os.get_app()
 
-)
+if __name__ == "__main__":
+    agent_os.serve(app="main:app", port=8000,reload=True)

prompts.py

@@ -0,0 +1,45 @@
+SHOPPING_AGENT_PROMPT = """
+<role>
+You are **Gem**, a personalized shopping AI Agent. Your goal is to help user find the best products based on their needs.
+</role>
+
+<task>
+
+1.  **Analyze the User Need**: Chat with the user to help them identify the product, along with any specific constraints like price, rating, or category.
+2.  **Fetch Products**: Once they have given you the details about the product they want, use the `fetch_products` tool to get the products. This tool will automatically display the products to the user in frontend. Follow these instructions when using the tool:
+        * **Extract `filters`**: Identify any explicit constraints and populate the `filters` object.
+            -   **Price**: Look for terms like "under $50", "over 100 dollars", "between $20 and $60".
+            -   **Rating**: Look for terms like "highly rated", "4 stars or more", "best reviewed". A generic term like "best" can imply a `min_rating` of `4.0`.
+            -   **Rating Count**: Look for terms like "popular", "most reviewed". A generic term like "popular" can imply a `min_reviews` of `100`.
+            -   **Categories**: Identify the categories from the user query. **MUST** be from the allowed categories listed below. Do not invent new categories. For example, if the user says "I need black shirts for men", then the right category would be "men's clothing".
+3.  **Suggest Products**: Analyze the retrieved products. Provide recommendations to the user. DO NOT write each product details again, because the products are already displayed in frontend. Instead, you should highlight the best ones based on the user's needs and give them personalized suggestions.
+
+</task>
+
+<allowed_categories>
+You must use one of the following category values if a category filter is applicable:
+- appliances
+- car & motorbike
+- tv, audio & cameras
+- sports & fitness
+- grocery & gourmet foods
+- home & kitchen
+- pet supplies
+- stores
+- toys & baby products
+- kids' fashion
+- bags & luggage
+- accessories
+- women's shoes
+- beauty & health
+- men's shoes
+- women's clothing
+- industrial supplies
+- men's clothing
+- music
+- home, kitchen, pets
+
+NOTE: These categories are case-sensitive. So you must use the exact category values provided.
+</allowed_categories>
+
+"""

pyproject.toml

@@ -5,6 +5,7 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [
+    "ag-ui-protocol>=0.1.9",
     "agno>=2.0.8",
     "cohere>=5.18.0",
     "ddgs>=9.6.0",
@@ -18,6 +19,7 @@ dependencies = [
     "pinecone>=7.3.0",
     "pinecone-client[grpc]>=6.0.0",
     "pinecone-text>=0.11.0",
+    "psycopg2>=2.9.10",
     "python-dotenv>=1.1.1",
     "uvicorn>=0.37.0",
 ]

retriever.py

@@ -152,8 +152,8 @@ class PineconeHybridRetriever:
             try:
                 product = ProductItem(
                     name=metadata['name'],
-                    price=metadata['discount_price_usd'],
-                    original_price=metadata['actual_price_usd'],
+                    price=round(metadata['discount_price_usd'], 2),
+                    original_price=round(metadata['actual_price_usd'], 2),
                     rating=metadata['ratings'],
                     num_reviews=metadata['no_of_ratings'],
                     category=metadata['main_category'],

tools.py

@@ -0,0 +1,33 @@
+from agno.tools import tool
+from retriever import PineconeHybridRetriever, FilterModel, ProductItem
+
+from typing import Optional
+import json
+from dotenv import load_dotenv
+
+load_dotenv()
+
+@tool
+def fetch_products(
+    query: str, limit: int = 5, filters: Optional[FilterModel] = None
+) -> list[dict]:
+    """
+    Fetch products from database based on query and filters
+
+    Args:
+        query (str): Search query
+        limit (int, optional): Number of products to return. Defaults to 5.
+        filters (FilterModel, optional): Filters to apply. Defaults to None.
+
+    Returns:
+        list[dict]: List of products dictionary items matching the query and filters
+    """
+    retriever = PineconeHybridRetriever("amazon-products-catalog")
+    products = retriever.search_products(
+        query=query,
+        filters=filters,
+        limit=limit,
+        use_hybrid_search=True,
+        enable_reranking=True,
+    )
+    return json.dumps([product.model_dump() for product in products])

uv.lock

@@ -7,6 +7,18 @@ resolution-markers = [
     "python_full_version >= '4'",
 ]
 
+[[package]]
+name = "ag-ui-protocol"
+version = "0.1.9"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7b/d7/a8f8789b3b8b5f7263a902361468e8dfefd85ec63d1d5398579b9175d76d/ag_ui_protocol-0.1.9.tar.gz", hash = "sha256:94d75e3919ff75e0b608a7eed445062ea0e6f11cd33b3386a7649047e0c7abd3", size = 4988, upload-time = "2025-09-19T13:36:26.903Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/39/50/2bb71a2a9135f4d88706293773320d185789b592987c09f79e9bf2f4875f/ag_ui_protocol-0.1.9-py3-none-any.whl", hash = "sha256:44c1238b0576a3915b3a16e1b3855724e08e92ebc96b1ff29379fbd3bfbd400b", size = 7070, upload-time = "2025-09-19T13:36:25.791Z" },
+]
+
 [[package]]
 name = "agno"
 version = "2.0.8"
@@ -1946,6 +1958,17 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/26/65/1070a6e3c036f39142c2820c4b52e9243246fcfc3f96239ac84472ba361e/psutil-7.1.0-cp37-abi3-win_arm64.whl", hash = "sha256:6937cb68133e7c97b6cc9649a570c9a18ba0efebed46d8c5dae4c07fa1b67a07", size = 244971, upload-time = "2025-09-17T20:15:12.262Z" },
 ]
 
+[[package]]
+name = "psycopg2"
+version = "2.9.10"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/62/51/2007ea29e605957a17ac6357115d0c1a1b60c8c984951c19419b3474cdfd/psycopg2-2.9.10.tar.gz", hash = "sha256:12ec0b40b0273f95296233e8750441339298e6a572f7039da5b260e3c8b60e11", size = 385672, upload-time = "2024-10-16T11:24:54.832Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/16/4623fad6076448df21c1a870c93a9774ad8a7b4dd1660223b59082dd8fec/psycopg2-2.9.10-cp312-cp312-win32.whl", hash = "sha256:65a63d7ab0e067e2cdb3cf266de39663203d38d6a8ed97f5ca0cb315c73fe067", size = 1025113, upload-time = "2024-10-16T11:18:40.148Z" },
+    { url = "https://files.pythonhosted.org/packages/66/de/baed128ae0fc07460d9399d82e631ea31a1f171c0c4ae18f9808ac6759e3/psycopg2-2.9.10-cp312-cp312-win_amd64.whl", hash = "sha256:4a579d6243da40a7b3182e0430493dbd55950c493d8c68f4eec0b302f6bbf20e", size = 1163951, upload-time = "2024-10-16T11:18:44.377Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/49/a6cfc94a9c483b1fa401fbcb23aca7892f60c7269c5ffa2ac408364f80dc/psycopg2-2.9.10-cp313-cp313-win_amd64.whl", hash = "sha256:91fd603a2155da8d0cfcdbf8ab24a2d54bca72795b90d2a3ed2b6da8d979dee2", size = 2569060, upload-time = "2025-01-04T20:09:15.28Z" },
+]
+
 [[package]]
 name = "ptyprocess"
 version = "0.7.0"
@@ -2420,6 +2443,7 @@ name = "shopping-ai-agent"
 version = "0.1.0"
 source = { virtual = "." }
 dependencies = [
+    { name = "ag-ui-protocol" },
     { name = "agno" },
     { name = "cohere" },
     { name = "ddgs" },
@@ -2433,12 +2457,14 @@ dependencies = [
     { name = "pinecone" },
     { name = "pinecone-client", extra = ["grpc"] },
     { name = "pinecone-text" },
+    { name = "psycopg2" },
     { name = "python-dotenv" },
     { name = "uvicorn" },
 ]
 
 [package.metadata]
 requires-dist = [
+    { name = "ag-ui-protocol", specifier = ">=0.1.9" },
     { name = "agno", specifier = ">=2.0.8" },
     { name = "cohere", specifier = ">=5.18.0" },
     { name = "ddgs", specifier = ">=9.6.0" },
@@ -2452,6 +2478,7 @@ requires-dist = [
     { name = "pinecone", specifier = ">=7.3.0" },
     { name = "pinecone-client", extras = ["grpc"], specifier = ">=6.0.0" },
     { name = "pinecone-text", specifier = ">=0.11.0" },
+    { name = "psycopg2", specifier = ">=2.9.10" },
     { name = "python-dotenv", specifier = ">=1.1.1" },
     { name = "uvicorn", specifier = ">=0.37.0" },
 ]