Feat: Pulumi S3

.gitignore

@@ -5,4 +5,4 @@ venv/
 .env
 .env.*
 pyproject.toml
-
+Makefile

back/app/backend/config.py

@@ -1,6 +1,5 @@
 import os
 
-TABLE = os.environ["TABLE"]
 REGION = os.environ["REGION"]
 AWS_ACCOUNT = os.environ["AWS_ACCOUNT"]
-SECRET_NAME = os.environ["SECRET_NAME"]
+KNOWLEDGE_BASE_ID = os.environ["KNOWLEDGE_BASE_ID"]

back/app/backend/dynamo.py

@@ -4,9 +4,8 @@ import os
 from botocore.exceptions import ClientError
 from langfuse import Langfuse
 
-from .config import REGION, TABLE, SECRET_NAME
+from .config import REGION, SECRET_NAME
 
-dynamodb = boto3.resource("dynamodb", region_name=REGION)
 
 
 def get_secret() -> str:
@@ -19,35 +18,9 @@ def get_secret() -> str:
     return response["SecretString"]
 
 
-secrets = json.loads(get_secret())
+#secrets = json.loads(get_secret())
-langfuse = Langfuse(
+#langfuse = Langfuse(
-    public_key=secrets["LANGFUSE-PUBLIC-KEY"],
+#    public_key=secrets["LANGFUSE-PUBLIC-KEY"],
-    secret_key=secrets["LANGFUSE-SECRET-KEY"],
+#    secret_key=secrets["LANGFUSE-SECRET-KEY"],
-    host=os.environ["LANGFUSE_HOST"],
+#    host=os.environ["LANGFUSE_HOST"],
-)
+#)
-
-
-def get_contexto(dashboard: str) -> dict:
-    """
-    Get contexto, filter, and items_disponiveis from DynamoDB for a given dashboard.
-
-    Returns:
-        Dict with 'contexto', 'filter', and 'items_disponiveis' keys
-    """
-    try:
-        table = dynamodb.Table(TABLE)
-        response = table.get_item(Key={"id": dashboard + "_contexto"})
-
-        if "Item" not in response:
-            return {"contexto": "", "filter": "", "items_disponiveis": {}}
-
-        item = response["Item"]
-        return {
-            "contexto": item.get("contexto", ""),
-            "filter": item.get("filter_key", ""),
-            "items_disponiveis": item.get("itens_disponiveis", {}),
-        }
-
-    except ClientError as e:
-        error_message = e.response["Error"]["Message"]
-        return {"contexto": f"Error: {error_message}", "filter": "", "items_disponiveis": {}}

back/app/backend/orquestrador.py

@@ -2,17 +2,16 @@ from langchain_core.messages import HumanMessage, AIMessage, SystemMessage
 from langfuse.langchain import CallbackHandler
 
 from .config import REGION
-from .dynamo import langfuse, get_contexto
 from .agent_bedrock import create_agent
-from .tools import ReportTools
+from .tools import build_knowledge_base_tool
 
 
-def main(user_query, history, model, base):
+def main(user_query, history, model="anthropic.claude-sonnet-4-5-20250929-v1:0"):
     """Main execution function."""
-    report_tools = []
+    report_tools = [build_knowledge_base_tool()]
 
 
-    SYSTEM_PROMPT = """"""
+    SYSTEM_PROMPT = """Você é um assistente de matrículas para o campus capivari do instituo federal de são paulo, tem acesso a uma tool que acessa uma knowledge base com informações sobre tanto a matricula dos alunos do técnico quanto superior do procedimento iterno, não responda perguntas sobre o meio de ingresso SISU."""
     langfuse_handler = CallbackHandler()
     agent = create_agent(model, REGION, tools=report_tools)
 
@@ -24,7 +23,7 @@ def main(user_query, history, model, base):
         "current_step": "init",
     }
 
-    config = {"callbacks": [langfuse_handler], "tags": [base]}
+    config = {"callbacks": [langfuse_handler]}
     final_state = agent.invoke(initial_state, config=config)
 
     total_input_tokens = 0
@@ -33,8 +32,6 @@ def main(user_query, history, model, base):
         if isinstance(msg, AIMessage) and hasattr(msg, "usage_metadata") and msg.usage_metadata:
             total_input_tokens += msg.usage_metadata.get("input_tokens", 0)
             total_output_tokens += msg.usage_metadata.get("output_tokens", 0)
-
-    langfuse.flush()
     return {
         "response": final_state["messages"][-1].content,
         "input_tokens": total_input_tokens,

back/app/backend/tools.py

@@ -1,85 +1,21 @@
-from botocore.exceptions import ClientError
+from langchain_aws import AmazonKnowledgeBasesRetriever
-from langchain_core.tools import StructuredTool
+from langchain_core.tools.retriever import create_retriever_tool
 
-from .config import TABLE
+from .config import REGION, KNOWLEDGE_BASE_ID
-from .dynamo import dynamodb
 
 
-class ReportTools:
+def build_knowledge_base_tool():
-    def __init__(self, id_mapping: dict[str, str]):
+    retriever = AmazonKnowledgeBasesRetriever(
-        self.id_mapping = id_mapping
+        knowledge_base_id=KNOWLEDGE_BASE_ID,
-
+        retrieval_config={"vectorSearchConfiguration": {"numberOfResults": 5}},
-    def get_variable_value(self, id: str, variable: str) -> str:
+        region_name=REGION,
-        """
+    )
-        Get a specific variable's value from DynamoDB for a specific id.
+    return create_retriever_tool(
-
+        retriever,
-        Args:
+        name="consultar_base_conhecimento",
-            id: The id of the data
+        description=(
-            variable: The variable/column name to retrieve from the table
+            "Consulta a base de conhecimento com informações sobre procedimentos internos de matrícula "
-
+            "para cursos técnicos e superiores do campus Capivari do IFSP. "
-        Returns:
+            "Use esta ferramenta para responder dúvidas sobre matrícula."
-            The content of the specified variable for the given id
+        ),
-        """
+    )
-        real_id = self.id_mapping.get(id, id)
-        try:
-            table = dynamodb.Table(TABLE)
-            response = table.get_item(Key={"id": real_id})
-
-            if "Item" not in response:
-                return f"No report found for month: {id}"
-
-            item = response["Item"]
-            content = item.get(variable, "")
-
-            if not content:
-                return f"Variable '{variable}' not found for month: {id}"
-
-            return f"<{id}>\n{content}\n</{id}>"
-
-        except ClientError as e:
-            error_message = e.response["Error"]["Message"]
-            return f"Error fetching report: {error_message}"
-
-    def get_variables_list(self, id: str) -> str:
-        """
-        Get the list of variables available in the table for a specific month.
-
-        Args:
-            id: The id of the data
-
-        Returns:
-            The list of available variables/keys for the specified data
-        """
-        real_id = self.id_mapping.get(id, id)
-        try:
-            table = dynamodb.Table(TABLE)
-            response = table.get_item(Key={"id": real_id})
-
-            if "Item" not in response:
-                return f"No data found for month: {id}"
-
-            item = response["Item"]
-            chaves_consolidadas = item.get("chaves_consolidadas", "")
-
-            if not chaves_consolidadas:
-                return f"No consolidated keys found for id: {id}"
-
-            return chaves_consolidadas
-
-        except ClientError as e:
-            error_message = e.response["Error"]["Message"]
-            return f"Error fetching consolidated keys: {error_message}"
-
-    def as_tools(self) -> list:
-        return [
-            StructuredTool.from_function(
-                self.get_variable_value,
-                name="get_variable_value",
-                description="Get a specific variable's data from DynamoDB for a specific id.",
-            ),
-            StructuredTool.from_function(
-                self.get_variables_list,
-                name="get_variable_list",
-                description="Get the list of variables available in the table for a specific id.",
-            ),
-        ]

back/app/front.py

@@ -41,7 +41,7 @@ if prompt := st.chat_input("Type your message here..."):
         
         # Simulate streaming response (replace with actual API call)
 
-        result = orquestrador.main(prompt,str(st.session_state.messages),selected_value,base)
+        result = orquestrador.main(prompt,str(st.session_state.messages))
         full_response = result["response"]
 
         # Simulate typing effect

back/requirements.txt

@@ -1,8 +1,8 @@
 boto3==1.42.10
-langchain-aws==1.1.0
+langchain-aws==1.3.0
-langgraph==1.0.5
+langgraph==1.0.9
 langchain==1.2.0
-streamlit==1.52.2
+streamlit==1.54.0
-langfuse==3.11.2
+langfuse==3.14.5
-fastapi==0.129.0
+fastapi==0.133.0
 uvicorn==0.41.0

infra/ecs_alb/Pulumi.ifsp-assistente-matricula-ecs-alb.yaml

@@ -1,32 +1,32 @@
 config:
   aws:region: us-east-1
   app-ecs:account_id: "305427701314" # dnxbrasil-nonprod
-  app-ecs:project_name: assistente-analitico
+  app-ecs:project_name: assistente-matricula
   app-ecs:environment: dev
   # app-ecs:bedrock_api_key: 
   #   secure: you-can-put-your-pulumi-encrypted-secure-string-here
   app-ecs:tags:
-    project: assistente-analitico-db-dev
+    project: assistente-matricula-dev
     env: dev                  # dev, test, stage, prod
     account: nonprod          # prod, nonprod, dataScience
     costCenter: AI            # AWSGeneral, AI, data, productName
     owner: AI                 # team or a preson responsible
   app-ecs:network:
-    vpc_id: vpc-17ceb96c
+    vpc_id: vpc-08b91683af59eab85
     alb_internal: false
     alb_subnet_ids: # 2+ private subnets if alb_internal else public subnets in the same region and vpc
-      - subnet-0de9f056635629827
+      - subnet-07886057270d8c0f3
-      - subnet-09cda74f27c543521
+      - subnet-0b40078e0a65d2a0c
     alb_allow_ingress_cidr:
       - 3.14.44.224/32
     ecs_subnet_ids:
-      - subnet-0f50f25a2fbb054d4
+      - subnet-07ccfbdb115f03d5e
-      - subnet-043a427630309c2f4
+      - subnet-0e05d1de8407fb798
   app-ecs:ecs:
-    - task_name: assisnte-analitico-db-dev
+    - task_name: assistente-matricula-dev
-      ecr_repo_name: assistente-analitico-db-dev
+      ecr_repo_name: assistente-matricula-dev
       ecr_image_tag: latest
-      ecr_image_digest: sha256:0bd3a927df4367ba29dbd173e0414d884e973c37599a3f6241341e8d190e827b 
+      ecr_image_digest: sha256:a4ff97f1bbc3ba14d40b32e68f6fc431dd5f05c70aff2f8b914f503e62bc7230 
       cpu: 256
       memory: 512
       desired_count: 1
@@ -46,14 +46,14 @@ config:
           target_port: 8000
           container_port: 8000
       env_variables:
-        LANGFUSE_HOST: http://172.31.252.176:3000
+        LANGFUSE_HOST: http://0.0.0.0/0
-        TABLE: poc_dnx_monthly_summary
         REGION: us-east-1
         AWS_ACCOUNT: "305427701314"
-        SECRET_NAME: assistente-db-secrets-manager
+        KNOWLEDGE_BASE_ID: LBH9H6QTAK
+        #SECRET_NAME: assistente-db-secrets-manager
         # SECRET_NAME: dev/ai-pge-doc-classification
         # BEDROCK_REGION: us-east-1
         # LANGCHAIN_TRACING_V2: "true"
         # LANGCHAIN_PROJECT: pge-doc-classification-dev
   app-ecs:cloudwatch:
-    log_group_name: assistente-analitico-db-dev
+    log_group_name: assistente-matricula-dev

infra/s3/Pulumi.assistente-matricula.yaml

@@ -1,9 +0,0 @@
-config:
-  aws:region: us-east-1
-  ia-bucket:
-    buckets:
-    - name: assistente-matricula-docs
-    - name: assistente-matricula-transcribe
-    tags:
-      Enviroment: dev
-      Owner: ai

infra/s3/Pulumi.ifsp-assistente-matricula.yaml

@@ -0,0 +1,21 @@
+config:
+  aws:region: us-east-1
+  ia-bucket:buckets:
+    - name: br-edu-ifsp-capivari-s3-docs-matricula-dev
+    - name: br-edu-ifsp-capivari-s3-transcribe-matricula-dev
+  ia-bucket:tags:
+      Nome: Bucket Knowledge Base
+      Ambiente: dev
+      projeto: Assistente matricula
+      responsavel: infra@ifsp.edu.br
+      centro-de-custo:  ti-geral-2026
+      criticidade: alta
+      data-de-criacao: 13/03/2026
+      backup: não
+      servico: Armazenamento
+      setor: ti
+      campus: capivari
+      area: infraestrutura
+      prospeccao: sim
+      poc: sim
+  

infra/s3/__main__.py

@@ -4,6 +4,6 @@ config=pulumi.Config()
 buckets_config=config.require_object("buckets")
 tags=config.require_object("tags")
 for b in buckets_config:
-    bucket=aws.s3.BucketV2(b["name"],tags=tags)
+    bucket=aws.s3.Bucket(b["name"],tags=tags)
     pulumi.export("bucket_name",bucket.id)
     pulumi.export("bucket_arn",bucket.arn)