Enhance RAG (Retrieval-Augmented Generation) functionality and dependencies

- Added RAG configuration management and integrated support for LangChain, ChromaDB, and sentence-transformers.
- Implemented document ingestion, chunking, and embedding management for RAG capabilities.
- Updated application to check and install necessary RAG dependencies automatically.
- Enhanced README to include RAG features and usage instructions.
- Refactored environment checks to ensure RAG dependencies are available.
- Improved UI to support document conversion and chat functionalities with RAG integration.
- Updated requirements.txt to include new RAG-related packages.

.gitignore

@@ -97,3 +97,6 @@ app_backup.py
 
 #Ignore .claude
 .claude/
+
+# Ignore data folder
+/data/

=0.1.0

@@ -0,0 +1,96 @@
+Collecting langchain-chroma
+  Using cached langchain_chroma-0.2.4-py3-none-any.whl.metadata (1.1 kB)
+Requirement already satisfied: langchain-core>=0.3.60 in ./.venv/lib/python3.12/site-packages (from langchain-chroma) (0.3.66)
+Requirement already satisfied: numpy>=1.26.0 in ./.venv/lib/python3.12/site-packages (from langchain-chroma) (1.26.3)
+Requirement already satisfied: chromadb>=1.0.9 in ./.venv/lib/python3.12/site-packages (from langchain-chroma) (1.0.13)
+Requirement already satisfied: build>=1.0.3 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.2.2.post1)
+Requirement already satisfied: pydantic>=1.9 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (2.11.7)
+Requirement already satisfied: pybase64>=1.4.1 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.4.1)
+Requirement already satisfied: uvicorn>=0.18.3 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (0.34.3)
+Requirement already satisfied: posthog>=2.4.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (5.4.0)
+Requirement already satisfied: typing-extensions>=4.5.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (4.14.0)
+Requirement already satisfied: onnxruntime>=1.14.1 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.22.0)
+Requirement already satisfied: opentelemetry-api>=1.2.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.34.1)
+Requirement already satisfied: opentelemetry-exporter-otlp-proto-grpc>=1.2.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.34.1)
+Requirement already satisfied: opentelemetry-sdk>=1.2.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.34.1)
+Requirement already satisfied: tokenizers>=0.13.2 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (0.21.1)
+Requirement already satisfied: pypika>=0.48.9 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (0.48.9)
+Requirement already satisfied: tqdm>=4.65.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (4.67.1)
+Requirement already satisfied: overrides>=7.3.1 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (7.7.0)
+Requirement already satisfied: importlib-resources in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (6.5.2)
+Requirement already satisfied: grpcio>=1.58.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (1.73.0)
+Requirement already satisfied: bcrypt>=4.0.1 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (4.3.0)
+Requirement already satisfied: typer>=0.9.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (0.16.0)
+Requirement already satisfied: kubernetes>=28.1.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (33.1.0)
+Requirement already satisfied: tenacity>=8.2.3 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (8.5.0)
+Requirement already satisfied: pyyaml>=6.0.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (6.0.2)
+Requirement already satisfied: mmh3>=4.0.1 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (5.1.0)
+Requirement already satisfied: orjson>=3.9.12 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (3.10.18)
+Requirement already satisfied: httpx>=0.27.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (0.28.1)
+Requirement already satisfied: rich>=10.11.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (14.0.0)
+Requirement already satisfied: jsonschema>=4.19.0 in ./.venv/lib/python3.12/site-packages (from chromadb>=1.0.9->langchain-chroma) (4.24.0)
+Requirement already satisfied: langsmith>=0.3.45 in ./.venv/lib/python3.12/site-packages (from langchain-core>=0.3.60->langchain-chroma) (0.4.1)
+Requirement already satisfied: jsonpatch<2.0,>=1.33 in ./.venv/lib/python3.12/site-packages (from langchain-core>=0.3.60->langchain-chroma) (1.33)
+Requirement already satisfied: packaging<25,>=23.2 in ./.venv/lib/python3.12/site-packages (from langchain-core>=0.3.60->langchain-chroma) (24.2)
+Requirement already satisfied: pyproject_hooks in ./.venv/lib/python3.12/site-packages (from build>=1.0.3->chromadb>=1.0.9->langchain-chroma) (1.2.0)
+Requirement already satisfied: anyio in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (4.9.0)
+Requirement already satisfied: certifi in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (2025.6.15)
+Requirement already satisfied: httpcore==1.* in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (1.0.9)
+Requirement already satisfied: idna in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (3.10)
+Requirement already satisfied: h11>=0.16 in ./.venv/lib/python3.12/site-packages (from httpcore==1.*->httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (0.16.0)
+Requirement already satisfied: jsonpointer>=1.9 in ./.venv/lib/python3.12/site-packages (from jsonpatch<2.0,>=1.33->langchain-core>=0.3.60->langchain-chroma) (3.0.0)
+Requirement already satisfied: attrs>=22.2.0 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb>=1.0.9->langchain-chroma) (25.3.0)
+Requirement already satisfied: jsonschema-specifications>=2023.03.6 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb>=1.0.9->langchain-chroma) (2025.4.1)
+Requirement already satisfied: referencing>=0.28.4 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb>=1.0.9->langchain-chroma) (0.36.2)
+Requirement already satisfied: rpds-py>=0.7.1 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb>=1.0.9->langchain-chroma) (0.25.1)
+Requirement already satisfied: six>=1.9.0 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (1.17.0)
+Requirement already satisfied: python-dateutil>=2.5.3 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (2.9.0.post0)
+Requirement already satisfied: google-auth>=1.0.1 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (2.40.3)
+Requirement already satisfied: websocket-client!=0.40.0,!=0.41.*,!=0.42.*,>=0.32.0 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (1.8.0)
+Requirement already satisfied: requests in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (2.32.4)
+Requirement already satisfied: requests-oauthlib in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (2.0.0)
+Requirement already satisfied: oauthlib>=3.2.2 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (3.3.1)
+Requirement already satisfied: urllib3>=1.24.2 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (2.5.0)
+Requirement already satisfied: durationpy>=0.7 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (0.10)
+Requirement already satisfied: requests-toolbelt<2.0.0,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core>=0.3.60->langchain-chroma) (1.0.0)
+Requirement already satisfied: zstandard<0.24.0,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core>=0.3.60->langchain-chroma) (0.23.0)
+Requirement already satisfied: coloredlogs in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (15.0.1)
+Requirement already satisfied: flatbuffers in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (25.2.10)
+Requirement already satisfied: protobuf in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (5.29.5)
+Requirement already satisfied: sympy in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (1.14.0)
+Requirement already satisfied: importlib-metadata<8.8.0,>=6.0 in ./.venv/lib/python3.12/site-packages (from opentelemetry-api>=1.2.0->chromadb>=1.0.9->langchain-chroma) (8.7.0)
+Requirement already satisfied: googleapis-common-protos~=1.52 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb>=1.0.9->langchain-chroma) (1.70.0)
+Requirement already satisfied: opentelemetry-exporter-otlp-proto-common==1.34.1 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb>=1.0.9->langchain-chroma) (1.34.1)
+Requirement already satisfied: opentelemetry-proto==1.34.1 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb>=1.0.9->langchain-chroma) (1.34.1)
+Requirement already satisfied: opentelemetry-semantic-conventions==0.55b1 in ./.venv/lib/python3.12/site-packages (from opentelemetry-sdk>=1.2.0->chromadb>=1.0.9->langchain-chroma) (0.55b1)
+Requirement already satisfied: backoff>=1.10.0 in ./.venv/lib/python3.12/site-packages (from posthog>=2.4.0->chromadb>=1.0.9->langchain-chroma) (2.2.1)
+Requirement already satisfied: distro>=1.5.0 in ./.venv/lib/python3.12/site-packages (from posthog>=2.4.0->chromadb>=1.0.9->langchain-chroma) (1.9.0)
+Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb>=1.0.9->langchain-chroma) (0.7.0)
+Requirement already satisfied: pydantic-core==2.33.2 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb>=1.0.9->langchain-chroma) (2.33.2)
+Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb>=1.0.9->langchain-chroma) (0.4.1)
+Requirement already satisfied: markdown-it-py>=2.2.0 in ./.venv/lib/python3.12/site-packages (from rich>=10.11.0->chromadb>=1.0.9->langchain-chroma) (3.0.0)
+Requirement already satisfied: pygments<3.0.0,>=2.13.0 in ./.venv/lib/python3.12/site-packages (from rich>=10.11.0->chromadb>=1.0.9->langchain-chroma) (2.19.2)
+Requirement already satisfied: huggingface-hub<1.0,>=0.16.4 in ./.venv/lib/python3.12/site-packages (from tokenizers>=0.13.2->chromadb>=1.0.9->langchain-chroma) (0.33.0)
+Requirement already satisfied: click>=8.0.0 in ./.venv/lib/python3.12/site-packages (from typer>=0.9.0->chromadb>=1.0.9->langchain-chroma) (8.2.1)
+Requirement already satisfied: shellingham>=1.3.0 in ./.venv/lib/python3.12/site-packages (from typer>=0.9.0->chromadb>=1.0.9->langchain-chroma) (1.5.4)
+Requirement already satisfied: httptools>=0.6.3 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (0.6.4)
+Requirement already satisfied: python-dotenv>=0.13 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (1.1.0)
+Requirement already satisfied: uvloop>=0.15.1 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (0.21.0)
+Requirement already satisfied: watchfiles>=0.13 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (1.1.0)
+Requirement already satisfied: websockets>=10.4 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb>=1.0.9->langchain-chroma) (15.0.1)
+Requirement already satisfied: cachetools<6.0,>=2.0.0 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (5.5.2)
+Requirement already satisfied: pyasn1-modules>=0.2.1 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (0.4.2)
+Requirement already satisfied: rsa<5,>=3.1.4 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (4.9.1)
+Requirement already satisfied: filelock in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb>=1.0.9->langchain-chroma) (3.18.0)
+Requirement already satisfied: fsspec>=2023.5.0 in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb>=1.0.9->langchain-chroma) (2025.5.1)
+Requirement already satisfied: hf-xet<2.0.0,>=1.1.2 in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb>=1.0.9->langchain-chroma) (1.1.5)
+Requirement already satisfied: zipp>=3.20 in ./.venv/lib/python3.12/site-packages (from importlib-metadata<8.8.0,>=6.0->opentelemetry-api>=1.2.0->chromadb>=1.0.9->langchain-chroma) (3.23.0)
+Requirement already satisfied: mdurl~=0.1 in ./.venv/lib/python3.12/site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->chromadb>=1.0.9->langchain-chroma) (0.1.2)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests->kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (3.4.2)
+Requirement already satisfied: sniffio>=1.1 in ./.venv/lib/python3.12/site-packages (from anyio->httpx>=0.27.0->chromadb>=1.0.9->langchain-chroma) (1.3.1)
+Requirement already satisfied: humanfriendly>=9.1 in ./.venv/lib/python3.12/site-packages (from coloredlogs->onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (10.0)
+Requirement already satisfied: mpmath<1.4,>=1.1.0 in ./.venv/lib/python3.12/site-packages (from sympy->onnxruntime>=1.14.1->chromadb>=1.0.9->langchain-chroma) (1.3.0)
+Requirement already satisfied: pyasn1<0.7.0,>=0.6.1 in ./.venv/lib/python3.12/site-packages (from pyasn1-modules>=0.2.1->google-auth>=1.0.1->kubernetes>=28.1.0->chromadb>=1.0.9->langchain-chroma) (0.6.1)
+Using cached langchain_chroma-0.2.4-py3-none-any.whl (11 kB)
+Installing collected packages: langchain-chroma
+Successfully installed langchain-chroma-0.2.4

=0.2.0

@@ -0,0 +1,36 @@
+Collecting langchain-openai
+  Using cached langchain_openai-0.3.25-py3-none-any.whl.metadata (2.3 kB)
+Requirement already satisfied: langchain-core<1.0.0,>=0.3.66 in ./.venv/lib/python3.12/site-packages (from langchain-openai) (0.3.66)
+Requirement already satisfied: openai<2.0.0,>=1.86.0 in ./.venv/lib/python3.12/site-packages (from langchain-openai) (1.90.0)
+Requirement already satisfied: tiktoken<1,>=0.7 in ./.venv/lib/python3.12/site-packages (from langchain-openai) (0.9.0)
+Requirement already satisfied: langsmith>=0.3.45 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (0.4.1)
+Requirement already satisfied: tenacity!=8.4.0,<10.0.0,>=8.1.0 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (8.5.0)
+Requirement already satisfied: jsonpatch<2.0,>=1.33 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (1.33)
+Requirement already satisfied: PyYAML>=5.3 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (6.0.2)
+Requirement already satisfied: packaging<25,>=23.2 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (24.2)
+Requirement already satisfied: typing-extensions>=4.7 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (4.14.0)
+Requirement already satisfied: pydantic>=2.7.4 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.66->langchain-openai) (2.11.7)
+Requirement already satisfied: anyio<5,>=3.5.0 in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (4.9.0)
+Requirement already satisfied: distro<2,>=1.7.0 in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (1.9.0)
+Requirement already satisfied: httpx<1,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (0.28.1)
+Requirement already satisfied: jiter<1,>=0.4.0 in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (0.10.0)
+Requirement already satisfied: sniffio in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (1.3.1)
+Requirement already satisfied: tqdm>4 in ./.venv/lib/python3.12/site-packages (from openai<2.0.0,>=1.86.0->langchain-openai) (4.67.1)
+Requirement already satisfied: regex>=2022.1.18 in ./.venv/lib/python3.12/site-packages (from tiktoken<1,>=0.7->langchain-openai) (2024.11.6)
+Requirement already satisfied: requests>=2.26.0 in ./.venv/lib/python3.12/site-packages (from tiktoken<1,>=0.7->langchain-openai) (2.32.4)
+Requirement already satisfied: idna>=2.8 in ./.venv/lib/python3.12/site-packages (from anyio<5,>=3.5.0->openai<2.0.0,>=1.86.0->langchain-openai) (3.10)
+Requirement already satisfied: certifi in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->openai<2.0.0,>=1.86.0->langchain-openai) (2025.6.15)
+Requirement already satisfied: httpcore==1.* in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->openai<2.0.0,>=1.86.0->langchain-openai) (1.0.9)
+Requirement already satisfied: h11>=0.16 in ./.venv/lib/python3.12/site-packages (from httpcore==1.*->httpx<1,>=0.23.0->openai<2.0.0,>=1.86.0->langchain-openai) (0.16.0)
+Requirement already satisfied: jsonpointer>=1.9 in ./.venv/lib/python3.12/site-packages (from jsonpatch<2.0,>=1.33->langchain-core<1.0.0,>=0.3.66->langchain-openai) (3.0.0)
+Requirement already satisfied: orjson<4.0.0,>=3.9.14 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.66->langchain-openai) (3.10.18)
+Requirement already satisfied: requests-toolbelt<2.0.0,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.66->langchain-openai) (1.0.0)
+Requirement already satisfied: zstandard<0.24.0,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.66->langchain-openai) (0.23.0)
+Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.66->langchain-openai) (0.7.0)
+Requirement already satisfied: pydantic-core==2.33.2 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.66->langchain-openai) (2.33.2)
+Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.66->langchain-openai) (0.4.1)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests>=2.26.0->tiktoken<1,>=0.7->langchain-openai) (3.4.2)
+Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests>=2.26.0->tiktoken<1,>=0.7->langchain-openai) (2.5.0)
+Using cached langchain_openai-0.3.25-py3-none-any.whl (69 kB)
+Installing collected packages: langchain-openai
+Successfully installed langchain-openai-0.3.25

=0.3.0

@@ -0,0 +1,26 @@
+Requirement already satisfied: langchain-text-splitters in ./.venv/lib/python3.12/site-packages (0.3.8)
+Requirement already satisfied: langchain-core<1.0.0,>=0.3.51 in ./.venv/lib/python3.12/site-packages (from langchain-text-splitters) (0.3.66)
+Requirement already satisfied: langsmith>=0.3.45 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.4.1)
+Requirement already satisfied: tenacity!=8.4.0,<10.0.0,>=8.1.0 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (8.5.0)
+Requirement already satisfied: jsonpatch<2.0,>=1.33 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (1.33)
+Requirement already satisfied: PyYAML>=5.3 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (6.0.2)
+Requirement already satisfied: packaging<25,>=23.2 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (24.2)
+Requirement already satisfied: typing-extensions>=4.7 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (4.14.0)
+Requirement already satisfied: pydantic>=2.7.4 in ./.venv/lib/python3.12/site-packages (from langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (2.11.7)
+Requirement already satisfied: jsonpointer>=1.9 in ./.venv/lib/python3.12/site-packages (from jsonpatch<2.0,>=1.33->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (3.0.0)
+Requirement already satisfied: httpx<1,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.28.1)
+Requirement already satisfied: orjson<4.0.0,>=3.9.14 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (3.10.18)
+Requirement already satisfied: requests<3,>=2 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (2.32.4)
+Requirement already satisfied: requests-toolbelt<2.0.0,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (1.0.0)
+Requirement already satisfied: zstandard<0.24.0,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.23.0)
+Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.7.0)
+Requirement already satisfied: pydantic-core==2.33.2 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (2.33.2)
+Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=2.7.4->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.4.1)
+Requirement already satisfied: anyio in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (4.9.0)
+Requirement already satisfied: certifi in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (2025.6.15)
+Requirement already satisfied: httpcore==1.* in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (1.0.9)
+Requirement already satisfied: idna in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (3.10)
+Requirement already satisfied: h11>=0.16 in ./.venv/lib/python3.12/site-packages (from httpcore==1.*->httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (0.16.0)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests<3,>=2->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (3.4.2)
+Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests<3,>=2->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (2.5.0)
+Requirement already satisfied: sniffio>=1.1 in ./.venv/lib/python3.12/site-packages (from anyio->httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<1.0.0,>=0.3.51->langchain-text-splitters) (1.3.1)

=0.5.0

@@ -0,0 +1,102 @@
+Collecting chromadb
+  Using cached chromadb-1.0.13-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.0 kB)
+Requirement already satisfied: build>=1.0.3 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.2.2.post1)
+Requirement already satisfied: pydantic>=1.9 in ./.venv/lib/python3.12/site-packages (from chromadb) (2.11.7)
+Requirement already satisfied: pybase64>=1.4.1 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.4.1)
+Requirement already satisfied: uvicorn>=0.18.3 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (0.34.3)
+Requirement already satisfied: numpy>=1.22.5 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.26.3)
+Requirement already satisfied: posthog>=2.4.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (5.4.0)
+Requirement already satisfied: typing-extensions>=4.5.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (4.14.0)
+Requirement already satisfied: onnxruntime>=1.14.1 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.22.0)
+Requirement already satisfied: opentelemetry-api>=1.2.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.34.1)
+Collecting opentelemetry-exporter-otlp-proto-grpc>=1.2.0 (from chromadb)
+  Using cached opentelemetry_exporter_otlp_proto_grpc-1.34.1-py3-none-any.whl.metadata (2.4 kB)
+Collecting opentelemetry-sdk>=1.2.0 (from chromadb)
+  Using cached opentelemetry_sdk-1.34.1-py3-none-any.whl.metadata (1.6 kB)
+Requirement already satisfied: tokenizers>=0.13.2 in ./.venv/lib/python3.12/site-packages (from chromadb) (0.21.1)
+Requirement already satisfied: pypika>=0.48.9 in ./.venv/lib/python3.12/site-packages (from chromadb) (0.48.9)
+Requirement already satisfied: tqdm>=4.65.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (4.67.1)
+Requirement already satisfied: overrides>=7.3.1 in ./.venv/lib/python3.12/site-packages (from chromadb) (7.7.0)
+Requirement already satisfied: importlib-resources in ./.venv/lib/python3.12/site-packages (from chromadb) (6.5.2)
+Requirement already satisfied: grpcio>=1.58.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (1.73.0)
+Requirement already satisfied: bcrypt>=4.0.1 in ./.venv/lib/python3.12/site-packages (from chromadb) (4.3.0)
+Requirement already satisfied: typer>=0.9.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (0.16.0)
+Requirement already satisfied: kubernetes>=28.1.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (33.1.0)
+Requirement already satisfied: tenacity>=8.2.3 in ./.venv/lib/python3.12/site-packages (from chromadb) (8.5.0)
+Requirement already satisfied: pyyaml>=6.0.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (6.0.2)
+Requirement already satisfied: mmh3>=4.0.1 in ./.venv/lib/python3.12/site-packages (from chromadb) (5.1.0)
+Requirement already satisfied: orjson>=3.9.12 in ./.venv/lib/python3.12/site-packages (from chromadb) (3.10.18)
+Requirement already satisfied: httpx>=0.27.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (0.28.1)
+Requirement already satisfied: rich>=10.11.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (14.0.0)
+Requirement already satisfied: jsonschema>=4.19.0 in ./.venv/lib/python3.12/site-packages (from chromadb) (4.24.0)
+Requirement already satisfied: packaging>=19.1 in ./.venv/lib/python3.12/site-packages (from build>=1.0.3->chromadb) (24.2)
+Requirement already satisfied: pyproject_hooks in ./.venv/lib/python3.12/site-packages (from build>=1.0.3->chromadb) (1.2.0)
+Requirement already satisfied: anyio in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb) (4.9.0)
+Requirement already satisfied: certifi in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb) (2025.6.15)
+Requirement already satisfied: httpcore==1.* in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb) (1.0.9)
+Requirement already satisfied: idna in ./.venv/lib/python3.12/site-packages (from httpx>=0.27.0->chromadb) (3.10)
+Requirement already satisfied: h11>=0.16 in ./.venv/lib/python3.12/site-packages (from httpcore==1.*->httpx>=0.27.0->chromadb) (0.16.0)
+Requirement already satisfied: attrs>=22.2.0 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb) (25.3.0)
+Requirement already satisfied: jsonschema-specifications>=2023.03.6 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb) (2025.4.1)
+Requirement already satisfied: referencing>=0.28.4 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb) (0.36.2)
+Requirement already satisfied: rpds-py>=0.7.1 in ./.venv/lib/python3.12/site-packages (from jsonschema>=4.19.0->chromadb) (0.25.1)
+Requirement already satisfied: six>=1.9.0 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (1.17.0)
+Requirement already satisfied: python-dateutil>=2.5.3 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (2.9.0.post0)
+Requirement already satisfied: google-auth>=1.0.1 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (2.40.3)
+Requirement already satisfied: websocket-client!=0.40.0,!=0.41.*,!=0.42.*,>=0.32.0 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (1.8.0)
+Requirement already satisfied: requests in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (2.32.4)
+Requirement already satisfied: requests-oauthlib in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (2.0.0)
+Requirement already satisfied: oauthlib>=3.2.2 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (3.3.1)
+Requirement already satisfied: urllib3>=1.24.2 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (2.5.0)
+Requirement already satisfied: durationpy>=0.7 in ./.venv/lib/python3.12/site-packages (from kubernetes>=28.1.0->chromadb) (0.10)
+Requirement already satisfied: coloredlogs in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb) (15.0.1)
+Requirement already satisfied: flatbuffers in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb) (25.2.10)
+Requirement already satisfied: protobuf in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb) (6.31.1)
+Requirement already satisfied: sympy in ./.venv/lib/python3.12/site-packages (from onnxruntime>=1.14.1->chromadb) (1.14.0)
+Requirement already satisfied: importlib-metadata<8.8.0,>=6.0 in ./.venv/lib/python3.12/site-packages (from opentelemetry-api>=1.2.0->chromadb) (8.7.0)
+Requirement already satisfied: googleapis-common-protos~=1.52 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb) (1.70.0)
+Requirement already satisfied: opentelemetry-exporter-otlp-proto-common==1.34.1 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb) (1.34.1)
+Requirement already satisfied: opentelemetry-proto==1.34.1 in ./.venv/lib/python3.12/site-packages (from opentelemetry-exporter-otlp-proto-grpc>=1.2.0->chromadb) (1.34.1)
+Collecting protobuf (from onnxruntime>=1.14.1->chromadb)
+  Using cached protobuf-5.29.5-cp38-abi3-manylinux2014_x86_64.whl.metadata (592 bytes)
+Collecting opentelemetry-semantic-conventions==0.55b1 (from opentelemetry-sdk>=1.2.0->chromadb)
+  Using cached opentelemetry_semantic_conventions-0.55b1-py3-none-any.whl.metadata (2.5 kB)
+Requirement already satisfied: backoff>=1.10.0 in ./.venv/lib/python3.12/site-packages (from posthog>=2.4.0->chromadb) (2.2.1)
+Requirement already satisfied: distro>=1.5.0 in ./.venv/lib/python3.12/site-packages (from posthog>=2.4.0->chromadb) (1.9.0)
+Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb) (0.7.0)
+Requirement already satisfied: pydantic-core==2.33.2 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb) (2.33.2)
+Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic>=1.9->chromadb) (0.4.1)
+Requirement already satisfied: markdown-it-py>=2.2.0 in ./.venv/lib/python3.12/site-packages (from rich>=10.11.0->chromadb) (3.0.0)
+Requirement already satisfied: pygments<3.0.0,>=2.13.0 in ./.venv/lib/python3.12/site-packages (from rich>=10.11.0->chromadb) (2.19.2)
+Requirement already satisfied: huggingface-hub<1.0,>=0.16.4 in ./.venv/lib/python3.12/site-packages (from tokenizers>=0.13.2->chromadb) (0.33.0)
+Requirement already satisfied: click>=8.0.0 in ./.venv/lib/python3.12/site-packages (from typer>=0.9.0->chromadb) (8.2.1)
+Requirement already satisfied: shellingham>=1.3.0 in ./.venv/lib/python3.12/site-packages (from typer>=0.9.0->chromadb) (1.5.4)
+Requirement already satisfied: httptools>=0.6.3 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (0.6.4)
+Requirement already satisfied: python-dotenv>=0.13 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (1.1.0)
+Requirement already satisfied: uvloop>=0.15.1 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (0.21.0)
+Requirement already satisfied: watchfiles>=0.13 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (1.1.0)
+Requirement already satisfied: websockets>=10.4 in ./.venv/lib/python3.12/site-packages (from uvicorn[standard]>=0.18.3->chromadb) (15.0.1)
+Requirement already satisfied: cachetools<6.0,>=2.0.0 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb) (5.5.2)
+Requirement already satisfied: pyasn1-modules>=0.2.1 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb) (0.4.2)
+Requirement already satisfied: rsa<5,>=3.1.4 in ./.venv/lib/python3.12/site-packages (from google-auth>=1.0.1->kubernetes>=28.1.0->chromadb) (4.9.1)
+Requirement already satisfied: filelock in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb) (3.18.0)
+Requirement already satisfied: fsspec>=2023.5.0 in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb) (2025.5.1)
+Requirement already satisfied: hf-xet<2.0.0,>=1.1.2 in ./.venv/lib/python3.12/site-packages (from huggingface-hub<1.0,>=0.16.4->tokenizers>=0.13.2->chromadb) (1.1.5)
+Requirement already satisfied: zipp>=3.20 in ./.venv/lib/python3.12/site-packages (from importlib-metadata<8.8.0,>=6.0->opentelemetry-api>=1.2.0->chromadb) (3.23.0)
+Requirement already satisfied: mdurl~=0.1 in ./.venv/lib/python3.12/site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->chromadb) (0.1.2)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests->kubernetes>=28.1.0->chromadb) (3.4.2)
+Requirement already satisfied: sniffio>=1.1 in ./.venv/lib/python3.12/site-packages (from anyio->httpx>=0.27.0->chromadb) (1.3.1)
+Requirement already satisfied: humanfriendly>=9.1 in ./.venv/lib/python3.12/site-packages (from coloredlogs->onnxruntime>=1.14.1->chromadb) (10.0)
+Requirement already satisfied: mpmath<1.4,>=1.1.0 in ./.venv/lib/python3.12/site-packages (from sympy->onnxruntime>=1.14.1->chromadb) (1.3.0)
+Requirement already satisfied: pyasn1<0.7.0,>=0.6.1 in ./.venv/lib/python3.12/site-packages (from pyasn1-modules>=0.2.1->google-auth>=1.0.1->kubernetes>=28.1.0->chromadb) (0.6.1)
+Using cached chromadb-1.0.13-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (19.3 MB)
+Using cached opentelemetry_exporter_otlp_proto_grpc-1.34.1-py3-none-any.whl (18 kB)
+Using cached opentelemetry_sdk-1.34.1-py3-none-any.whl (118 kB)
+Using cached opentelemetry_semantic_conventions-0.55b1-py3-none-any.whl (196 kB)
+Using cached protobuf-5.29.5-cp38-abi3-manylinux2014_x86_64.whl (319 kB)
+Installing collected packages: protobuf, opentelemetry-semantic-conventions, opentelemetry-sdk, opentelemetry-exporter-otlp-proto-grpc, chromadb
+  Attempting uninstall: protobuf
+    Found existing installation: protobuf 6.31.1
+    Uninstalling protobuf-6.31.1:
+      Successfully uninstalled protobuf-6.31.1
+Successfully installed chromadb-1.0.13 opentelemetry-exporter-otlp-proto-grpc-1.34.1 opentelemetry-sdk-1.34.1 opentelemetry-semantic-conventions-0.55b1 protobuf-5.29.5

=2.0.0

@@ -0,0 +1,56 @@
+Collecting langchain-google-genai
+  Using cached langchain_google_genai-2.1.5-py3-none-any.whl.metadata (5.2 kB)
+Requirement already satisfied: filetype<2.0.0,>=1.2.0 in ./.venv/lib/python3.12/site-packages (from langchain-google-genai) (1.2.0)
+Collecting google-ai-generativelanguage<0.7.0,>=0.6.18 (from langchain-google-genai)
+  Using cached google_ai_generativelanguage-0.6.18-py3-none-any.whl.metadata (9.8 kB)
+Requirement already satisfied: langchain-core<0.4.0,>=0.3.62 in ./.venv/lib/python3.12/site-packages (from langchain-google-genai) (0.3.66)
+Requirement already satisfied: pydantic<3,>=2 in ./.venv/lib/python3.12/site-packages (from langchain-google-genai) (2.11.7)
+Collecting google-api-core!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1 (from google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai)
+  Using cached google_api_core-2.25.1-py3-none-any.whl.metadata (3.0 kB)
+Requirement already satisfied: google-auth!=2.24.0,!=2.25.0,<3.0.0,>=2.14.1 in ./.venv/lib/python3.12/site-packages (from google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (2.40.3)
+Requirement already satisfied: proto-plus<2.0.0,>=1.22.3 in ./.venv/lib/python3.12/site-packages (from google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (1.26.1)
+Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<7.0.0,>=3.20.2 in ./.venv/lib/python3.12/site-packages (from google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (5.29.5)
+Requirement already satisfied: langsmith>=0.3.45 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (0.4.1)
+Requirement already satisfied: tenacity!=8.4.0,<10.0.0,>=8.1.0 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (8.5.0)
+Requirement already satisfied: jsonpatch<2.0,>=1.33 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (1.33)
+Requirement already satisfied: PyYAML>=5.3 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (6.0.2)
+Requirement already satisfied: packaging<25,>=23.2 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (24.2)
+Requirement already satisfied: typing-extensions>=4.7 in ./.venv/lib/python3.12/site-packages (from langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (4.14.0)
+Requirement already satisfied: annotated-types>=0.6.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3,>=2->langchain-google-genai) (0.7.0)
+Requirement already satisfied: pydantic-core==2.33.2 in ./.venv/lib/python3.12/site-packages (from pydantic<3,>=2->langchain-google-genai) (2.33.2)
+Requirement already satisfied: typing-inspection>=0.4.0 in ./.venv/lib/python3.12/site-packages (from pydantic<3,>=2->langchain-google-genai) (0.4.1)
+Requirement already satisfied: googleapis-common-protos<2.0.0,>=1.56.2 in ./.venv/lib/python3.12/site-packages (from google-api-core!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (1.70.0)
+Requirement already satisfied: requests<3.0.0,>=2.18.0 in ./.venv/lib/python3.12/site-packages (from google-api-core!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (2.32.4)
+Requirement already satisfied: grpcio<2.0.0,>=1.33.2 in ./.venv/lib/python3.12/site-packages (from google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (1.73.0)
+Collecting grpcio-status<2.0.0,>=1.33.2 (from google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai)
+  Using cached grpcio_status-1.73.0-py3-none-any.whl.metadata (1.1 kB)
+Requirement already satisfied: cachetools<6.0,>=2.0.0 in ./.venv/lib/python3.12/site-packages (from google-auth!=2.24.0,!=2.25.0,<3.0.0,>=2.14.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (5.5.2)
+Requirement already satisfied: pyasn1-modules>=0.2.1 in ./.venv/lib/python3.12/site-packages (from google-auth!=2.24.0,!=2.25.0,<3.0.0,>=2.14.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (0.4.2)
+Requirement already satisfied: rsa<5,>=3.1.4 in ./.venv/lib/python3.12/site-packages (from google-auth!=2.24.0,!=2.25.0,<3.0.0,>=2.14.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (4.9.1)
+Requirement already satisfied: jsonpointer>=1.9 in ./.venv/lib/python3.12/site-packages (from jsonpatch<2.0,>=1.33->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (3.0.0)
+Requirement already satisfied: httpx<1,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (0.28.1)
+Requirement already satisfied: orjson<4.0.0,>=3.9.14 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (3.10.18)
+Requirement already satisfied: requests-toolbelt<2.0.0,>=1.0.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (1.0.0)
+Requirement already satisfied: zstandard<0.24.0,>=0.23.0 in ./.venv/lib/python3.12/site-packages (from langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (0.23.0)
+Collecting protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<7.0.0,>=3.20.2 (from google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai)
+  Using cached protobuf-6.31.1-cp39-abi3-manylinux2014_x86_64.whl.metadata (593 bytes)
+Requirement already satisfied: anyio in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (4.9.0)
+Requirement already satisfied: certifi in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (2025.6.15)
+Requirement already satisfied: httpcore==1.* in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (1.0.9)
+Requirement already satisfied: idna in ./.venv/lib/python3.12/site-packages (from httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (3.10)
+Requirement already satisfied: h11>=0.16 in ./.venv/lib/python3.12/site-packages (from httpcore==1.*->httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (0.16.0)
+Requirement already satisfied: pyasn1<0.7.0,>=0.6.1 in ./.venv/lib/python3.12/site-packages (from pyasn1-modules>=0.2.1->google-auth!=2.24.0,!=2.25.0,<3.0.0,>=2.14.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (0.6.1)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests<3.0.0,>=2.18.0->google-api-core!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (3.4.2)
+Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests<3.0.0,>=2.18.0->google-api-core!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0,>=1.34.1->google-ai-generativelanguage<0.7.0,>=0.6.18->langchain-google-genai) (2.5.0)
+Requirement already satisfied: sniffio>=1.1 in ./.venv/lib/python3.12/site-packages (from anyio->httpx<1,>=0.23.0->langsmith>=0.3.45->langchain-core<0.4.0,>=0.3.62->langchain-google-genai) (1.3.1)
+Using cached langchain_google_genai-2.1.5-py3-none-any.whl (44 kB)
+Using cached google_ai_generativelanguage-0.6.18-py3-none-any.whl (1.4 MB)
+Using cached google_api_core-2.25.1-py3-none-any.whl (160 kB)
+Downloading grpcio_status-1.73.0-py3-none-any.whl (14 kB)
+Using cached protobuf-6.31.1-cp39-abi3-manylinux2014_x86_64.whl (321 kB)
+Installing collected packages: protobuf, grpcio-status, google-api-core, google-ai-generativelanguage, langchain-google-genai
+  Attempting uninstall: protobuf
+    Found existing installation: protobuf 5.29.5
+    Uninstalling protobuf-5.29.5:
+      Successfully uninstalled protobuf-5.29.5
+Successfully installed google-ai-generativelanguage-0.6.18 google-api-core-2.25.1 grpcio-status-1.73.0 langchain-google-genai-2.1.5 protobuf-6.31.1

=3.0.0

@@ -0,0 +1,50 @@
+Collecting sentence-transformers
+  Using cached sentence_transformers-4.1.0-py3-none-any.whl.metadata (13 kB)
+Requirement already satisfied: transformers<5.0.0,>=4.41.0 in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (4.53.0.dev0)
+Requirement already satisfied: tqdm in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (4.67.1)
+Requirement already satisfied: torch>=1.11.0 in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (2.7.1)
+Requirement already satisfied: scikit-learn in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (1.7.0)
+Requirement already satisfied: scipy in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (1.16.0)
+Requirement already satisfied: huggingface-hub>=0.20.0 in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (0.33.0)
+Requirement already satisfied: Pillow in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (11.2.1)
+Requirement already satisfied: typing_extensions>=4.5.0 in ./.venv/lib/python3.12/site-packages (from sentence-transformers) (4.14.0)
+Requirement already satisfied: filelock in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (3.18.0)
+Requirement already satisfied: fsspec>=2023.5.0 in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (2025.5.1)
+Requirement already satisfied: packaging>=20.9 in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (24.2)
+Requirement already satisfied: pyyaml>=5.1 in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (6.0.2)
+Requirement already satisfied: requests in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (2.32.4)
+Requirement already satisfied: hf-xet<2.0.0,>=1.1.2 in ./.venv/lib/python3.12/site-packages (from huggingface-hub>=0.20.0->sentence-transformers) (1.1.5)
+Requirement already satisfied: setuptools in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (80.9.0)
+Requirement already satisfied: sympy>=1.13.3 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (1.14.0)
+Requirement already satisfied: networkx in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (3.5)
+Requirement already satisfied: jinja2 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (3.1.6)
+Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.6.77 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.77)
+Requirement already satisfied: nvidia-cuda-runtime-cu12==12.6.77 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.77)
+Requirement already satisfied: nvidia-cuda-cupti-cu12==12.6.80 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.80)
+Requirement already satisfied: nvidia-cudnn-cu12==9.5.1.17 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (9.5.1.17)
+Requirement already satisfied: nvidia-cublas-cu12==12.6.4.1 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.4.1)
+Requirement already satisfied: nvidia-cufft-cu12==11.3.0.4 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (11.3.0.4)
+Requirement already satisfied: nvidia-curand-cu12==10.3.7.77 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (10.3.7.77)
+Requirement already satisfied: nvidia-cusolver-cu12==11.7.1.2 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (11.7.1.2)
+Requirement already satisfied: nvidia-cusparse-cu12==12.5.4.2 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.5.4.2)
+Requirement already satisfied: nvidia-cusparselt-cu12==0.6.3 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (0.6.3)
+Requirement already satisfied: nvidia-nccl-cu12==2.26.2 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (2.26.2)
+Requirement already satisfied: nvidia-nvtx-cu12==12.6.77 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.77)
+Requirement already satisfied: nvidia-nvjitlink-cu12==12.6.85 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (12.6.85)
+Requirement already satisfied: nvidia-cufile-cu12==1.11.1.6 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (1.11.1.6)
+Requirement already satisfied: triton==3.3.1 in ./.venv/lib/python3.12/site-packages (from torch>=1.11.0->sentence-transformers) (3.3.1)
+Requirement already satisfied: numpy>=1.17 in ./.venv/lib/python3.12/site-packages (from transformers<5.0.0,>=4.41.0->sentence-transformers) (1.26.3)
+Requirement already satisfied: regex!=2019.12.17 in ./.venv/lib/python3.12/site-packages (from transformers<5.0.0,>=4.41.0->sentence-transformers) (2024.11.6)
+Requirement already satisfied: tokenizers<0.22,>=0.21 in ./.venv/lib/python3.12/site-packages (from transformers<5.0.0,>=4.41.0->sentence-transformers) (0.21.1)
+Requirement already satisfied: safetensors>=0.4.3 in ./.venv/lib/python3.12/site-packages (from transformers<5.0.0,>=4.41.0->sentence-transformers) (0.5.3)
+Requirement already satisfied: joblib>=1.2.0 in ./.venv/lib/python3.12/site-packages (from scikit-learn->sentence-transformers) (1.5.1)
+Requirement already satisfied: threadpoolctl>=3.1.0 in ./.venv/lib/python3.12/site-packages (from scikit-learn->sentence-transformers) (3.6.0)
+Requirement already satisfied: mpmath<1.4,>=1.1.0 in ./.venv/lib/python3.12/site-packages (from sympy>=1.13.3->torch>=1.11.0->sentence-transformers) (1.3.0)
+Requirement already satisfied: MarkupSafe>=2.0 in ./.venv/lib/python3.12/site-packages (from jinja2->torch>=1.11.0->sentence-transformers) (3.0.2)
+Requirement already satisfied: charset_normalizer<4,>=2 in ./.venv/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (3.4.2)
+Requirement already satisfied: idna<4,>=2.5 in ./.venv/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (3.10)
+Requirement already satisfied: urllib3<3,>=1.21.1 in ./.venv/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (2.5.0)
+Requirement already satisfied: certifi>=2017.4.17 in ./.venv/lib/python3.12/site-packages (from requests->huggingface-hub>=0.20.0->sentence-transformers) (2025.6.15)
+Using cached sentence_transformers-4.1.0-py3-none-any.whl (345 kB)
+Installing collected packages: sentence-transformers
+Successfully installed sentence-transformers-4.1.0

README.md

@@ -12,12 +12,13 @@ pinned: false
 hf_oauth: true
 ---
 
-# Document to Markdown Converter
+# Document to Markdown Converter with RAG Chat
 
-A Hugging Face Space that converts various document formats to Markdown, now with MarkItDown integration!
+A Hugging Face Space that converts various document formats to Markdown and lets you chat with your documents using RAG (Retrieval-Augmented Generation)!
 
-## Features
+## ✨ Key Features
 
+### Document Conversion
 - Convert PDFs, Office documents, images, and more to Markdown
 - Multiple parser options:
   - MarkItDown: For comprehensive document conversion
@@ -25,6 +26,21 @@ A Hugging Face Space that converts various document formats to Markdown, now wit
   - GOT-OCR: For image-based OCR with LaTeX support
   - Gemini Flash: For AI-powered text extraction
 - Download converted documents as Markdown files
+
+### 🤖 RAG Chat with Documents
+- **Chat with your converted documents** using advanced AI
+- **Intelligent document retrieval** using vector embeddings
+- **Markdown-aware chunking** that preserves tables and code blocks
+- **Streaming chat responses** for real-time interaction
+- **Chat history management** with session persistence
+- **Usage limits** to prevent abuse on public spaces
+- **Powered by Gemini 2.5 Flash** for high-quality responses
+- **OpenAI embeddings** for accurate document retrieval
+
+### User Interface
+- **Dual-tab interface**: Document Converter + Chat
+- **Real-time status monitoring** for RAG system
+- **Auto-ingestion** of converted documents into chat system
 - Clean, responsive UI
 
 ## Using MarkItDown & Docling
@@ -54,8 +70,8 @@ This app integrates multiple powerful document conversion libraries:
 The application uses centralized configuration management. You can enhance functionality by setting these environment variables:
 
 ### 🔑 **API Keys:**
-- `GOOGLE_API_KEY`: Used for Gemini Flash parser and LaTeX to Markdown conversion
-- `OPENAI_API_KEY`: Enables AI-based image descriptions in MarkItDown
+- `GOOGLE_API_KEY`: Used for Gemini Flash parser, LaTeX conversion, and **RAG chat functionality**
+- `OPENAI_API_KEY`: Enables AI-based image descriptions in MarkItDown and **vector embeddings for RAG**
 - `MISTRAL_API_KEY`: For Mistral OCR parser (if available)
 
 ### ⚙️ **Configuration Options:**
@@ -82,15 +98,39 @@ The application uses centralized configuration management. You can enhance funct
 - `MODEL_TEMPERATURE`: Model temperature for AI responses (default: 0.1)
 - `MODEL_MAX_TOKENS`: Maximum tokens for AI responses (default: 4096)
 
+### 🧠 **RAG Configuration:**
+- `VECTOR_STORE_PATH`: Path for vector database storage (default: ./data/vector_store)
+- `CHAT_HISTORY_PATH`: Path for chat history storage (default: ./data/chat_history)
+- `EMBEDDING_MODEL`: OpenAI embedding model (default: text-embedding-3-small)
+- `CHUNK_SIZE`: Document chunk size for RAG (default: 1000)
+- `CHUNK_OVERLAP`: Overlap between chunks (default: 200)
+- `MAX_MESSAGES_PER_SESSION`: Chat limit per session (default: 50)
+- `MAX_MESSAGES_PER_HOUR`: Chat limit per hour (default: 100)
+- `RETRIEVAL_K`: Number of documents to retrieve (default: 4)
+- `RAG_MODEL`: Model for RAG chat (default: gemini-2.5-flash)
+- `RAG_TEMPERATURE`: Temperature for RAG responses (default: 0.1)
+- `RAG_MAX_TOKENS`: Max tokens for RAG responses (default: 4096)
+
 ## Usage
 
-1. Select a file to upload
-2. Choose your preferred parser:
+### Document Conversion
+1. Go to the **"Document Converter"** tab
+2. Select a file to upload
+3. Choose your preferred parser:
    - **"MarkItDown"** for comprehensive document conversion
    - **"Docling"** for advanced PDF understanding and table extraction
-3. Select an OCR method based on your chosen parser
-4. Click "Convert"
-5. View the Markdown output and download the converted file
+4. Select an OCR method based on your chosen parser
+5. Click "Convert"
+6. View the Markdown output and download the converted file
+7. **Documents are automatically added to the RAG system** for chat functionality
+
+### 🤖 Chat with Documents
+1. Go to the **"Chat with Documents"** tab
+2. Check the system status to ensure RAG components are ready
+3. Ask questions about your converted documents
+4. Enjoy real-time streaming responses with document context
+5. Use "New Session" to start fresh conversations
+6. Monitor your usage limits in the status panel
 
 ## Local Development
 
@@ -102,6 +142,11 @@ The application uses centralized configuration management. You can enhance funct
    OPENAI_API_KEY=your_openai_api_key_here
    MISTRAL_API_KEY=your_mistral_api_key_here
    DEBUG=true
+   
+   # RAG Configuration (optional - uses defaults if not set)
+   MAX_MESSAGES_PER_SESSION=50
+   MAX_MESSAGES_PER_HOUR=100
+   CHUNK_SIZE=1000
    ```
 3. Install dependencies:
    ```bash
@@ -334,7 +379,7 @@ markit_v2/
 │   ├── main.py             # Application launcher
 │   ├── core/               # Core functionality and utilities
 │   │   ├── __init__.py     # Package initialization
-│   │   ├── config.py       # 🆕 Centralized configuration management
+│   │   ├── config.py       # 🆕 Centralized configuration management (with RAG settings)
 │   │   ├── exceptions.py   # 🆕 Custom exception hierarchy
 │   │   ├── logging_config.py # 🆕 Centralized logging setup
 │   │   ├── environment.py  # 🆕 Environment setup and dependency management
@@ -353,9 +398,17 @@ markit_v2/
 │   │   ├── got_ocr_parser.py # GOT-OCR parser for images
 │   │   ├── mistral_ocr_parser.py # 🆕 Mistral OCR parser
 │   │   └── gemini_flash_parser.py # Gemini Flash parser
+│   ├── rag/                # 🆕 RAG (Retrieval-Augmented Generation) system
+│   │   ├── __init__.py     # Package initialization
+│   │   ├── embeddings.py   # OpenAI embedding model management
+│   │   ├── chunking.py     # Markdown-aware document chunking
+│   │   ├── vector_store.py # Chroma vector database management
+│   │   ├── memory.py       # Chat history and session management
+│   │   ├── chat_service.py # RAG chat service with Gemini 2.5 Flash
+│   │   └── ingestion.py    # Document ingestion pipeline
 │   └── ui/                 # User interface layer
 │       ├── __init__.py     # Package initialization
-│       └── ui.py           # Gradio UI with enhanced error handling
+│       └── ui.py           # Gradio UI with dual tabs (Converter + Chat)
 ├── documents/              # Documentation and examples (gitignored)
 ├── tessdata/               # Tesseract OCR data (gitignored)
 └── tests/                  # Tests (future)
@@ -370,6 +423,17 @@ markit_v2/
 - **Enhanced Parser Interface**: Validation, metadata, and cancellation support
 - **Lightweight Launcher**: Quick development startup with `run_app.py`
 - **Centralized Logging**: Configurable logging system (`src/core/logging_config.py`)
+- **🆕 RAG System**: Complete RAG implementation with vector search and chat capabilities
+
+### 🧠 **RAG System Architecture:**
+- **Embeddings Management** (`src/rag/embeddings.py`): OpenAI text-embedding-3-small integration
+- **Markdown-Aware Chunking** (`src/rag/chunking.py`): Preserves tables and code blocks as whole units
+- **Vector Store** (`src/rag/vector_store.py`): Chroma database with persistent storage
+- **Chat Memory** (`src/rag/memory.py`): Session management and conversation history
+- **Chat Service** (`src/rag/chat_service.py`): Streaming RAG responses with Gemini 2.5 Flash
+- **Document Ingestion** (`src/rag/ingestion.py`): Automated pipeline for converting documents to RAG-ready format
+- **Usage Limiting**: Anti-abuse measures for public deployment
+- **Auto-Ingestion**: Seamless integration with document conversion workflow
 
 ### ZeroGPU Integration Notes
 

app.py

@@ -46,6 +46,24 @@ except ImportError as e:
     except ImportError:
         print("Installing Docling...")
         subprocess.run([sys.executable, "-m", "pip", "install", "-q", "docling"], check=False)
+    
+    # Check RAG dependencies as fallback
+    try:
+        from langchain_openai import OpenAIEmbeddings
+        print("RAG dependencies are available")
+    except ImportError:
+        print("Installing RAG dependencies...")
+        rag_packages = [
+            "langchain>=0.3.0",
+            "langchain-openai>=0.2.0", 
+            "langchain-google-genai>=2.0.0",
+            "langchain-chroma>=0.1.0",
+            "langchain-text-splitters>=0.3.0",
+            "chromadb>=0.5.0",
+            "sentence-transformers>=3.0.0"
+        ]
+        for package in rag_packages:
+            subprocess.run([sys.executable, "-m", "pip", "install", "-q", package], check=False)
 
 # Import main function with fallback strategies (HF Spaces compatibility)
 try:

requirements.txt

@@ -33,4 +33,13 @@ markitdown[all]
 openai>=1.1.0  # For LLM image description support
 
 # Docling dependencies
-docling
+docling
+
+# RAG and LangChain dependencies
+langchain>=0.3.0
+langchain-openai>=0.2.0
+langchain-google-genai>=2.0.0
+langchain-chroma>=0.1.0
+langchain-text-splitters>=0.3.0
+chromadb>=0.5.0
+sentence-transformers>=3.0.0

setup.sh

@@ -14,7 +14,8 @@ if [ "$EUID" -eq 0 ]; then
     echo "Installing system dependencies..."
     apt-get update && apt-get install -y \
         wget \
-        pkg-config
+        pkg-config \
+        ffmpeg
     echo "System dependencies installed successfully"
 else
     echo "Not running as root. Skipping system dependencies installation."
@@ -56,6 +57,17 @@ echo "Installing Docling..."
 pip install -q -U docling
 echo "Docling installed successfully"
 
+# Install LangChain and RAG dependencies
+echo "Installing LangChain and RAG dependencies..."
+pip install -q -U langchain>=0.3.0
+pip install -q -U langchain-openai>=0.2.0
+pip install -q -U langchain-google-genai>=2.0.0
+pip install -q -U langchain-chroma>=0.1.0
+pip install -q -U langchain-text-splitters>=0.3.0
+pip install -q -U chromadb>=0.5.0
+pip install -q -U sentence-transformers>=3.0.0
+echo "LangChain and RAG dependencies installed successfully"
+
 # Install the project in development mode only if setup.py or pyproject.toml exists
 if [ -f "setup.py" ] || [ -f "pyproject.toml" ]; then
     echo "Installing project in development mode..."

src/core/config.py

@@ -75,6 +75,62 @@ class DoclingConfig:
         self.ocr_cpu_threads = int(os.getenv("OMP_NUM_THREADS", self.ocr_cpu_threads))
 
 
+@dataclass
+class RAGConfig:
+    """Configuration for RAG (Retrieval-Augmented Generation) functionality."""
+    # Vector store settings
+    vector_store_path: str = "./data/vector_store"
+    collection_name: str = "markit_documents"
+    
+    # Chat history settings
+    chat_history_path: str = "./data/chat_history"
+    
+    # Embedding settings
+    embedding_model: str = "text-embedding-3-small"
+    embedding_chunk_size: int = 1000
+    
+    # Chunking settings
+    chunk_size: int = 1000
+    chunk_overlap: int = 200
+    
+    # Chat limits
+    max_messages_per_session: int = 50
+    max_messages_per_hour: int = 100
+    
+    # Retrieval settings
+    retrieval_k: int = 4
+    retrieval_score_threshold: float = 0.5
+    
+    # LLM settings for RAG
+    rag_model: str = "gemini-2.5-flash"
+    rag_temperature: float = 0.1
+    rag_max_tokens: int = 4096
+    
+    def __post_init__(self):
+        """Load RAG configuration from environment variables."""
+        # For HF Spaces, ensure data directories are created
+        if os.getenv("SPACE_ID"):  # HF Spaces environment
+            base_data_path = "/tmp/data" if not os.access("./data", os.W_OK) else "./data"
+            self.vector_store_path = os.getenv("VECTOR_STORE_PATH", f"{base_data_path}/vector_store")
+            self.chat_history_path = os.getenv("CHAT_HISTORY_PATH", f"{base_data_path}/chat_history")
+        else:
+            self.vector_store_path = os.getenv("VECTOR_STORE_PATH", self.vector_store_path)
+            self.chat_history_path = os.getenv("CHAT_HISTORY_PATH", self.chat_history_path)
+        
+        self.collection_name = os.getenv("VECTOR_STORE_COLLECTION", self.collection_name)
+        self.embedding_model = os.getenv("EMBEDDING_MODEL", self.embedding_model)
+        self.embedding_chunk_size = int(os.getenv("EMBEDDING_CHUNK_SIZE", self.embedding_chunk_size))
+        self.chunk_size = int(os.getenv("CHUNK_SIZE", self.chunk_size))
+        self.chunk_overlap = int(os.getenv("CHUNK_OVERLAP", self.chunk_overlap))
+        self.max_messages_per_session = int(os.getenv("MAX_MESSAGES_PER_SESSION", self.max_messages_per_session))
+        self.max_messages_per_hour = int(os.getenv("MAX_MESSAGES_PER_HOUR", self.max_messages_per_hour))
+        self.retrieval_k = int(os.getenv("RETRIEVAL_K", self.retrieval_k))
+        self.retrieval_score_threshold = float(os.getenv("RETRIEVAL_SCORE_THRESHOLD", self.retrieval_score_threshold))
+        self.rag_model = os.getenv("RAG_MODEL", self.rag_model)
+        self.rag_temperature = float(os.getenv("RAG_TEMPERATURE", self.rag_temperature))
+        self.rag_max_tokens = int(os.getenv("RAG_MAX_TOKENS", self.rag_max_tokens))
+
+
 @dataclass
 class AppConfig:
     """Main application configuration."""
@@ -99,6 +155,7 @@ class Config:
         self.model = ModelConfig()
         self.docling = DoclingConfig()
         self.app = AppConfig()
+        self.rag = RAGConfig()
     
     def validate(self) -> Dict[str, Any]:
         """Validate configuration and return validation results."""
@@ -115,6 +172,13 @@ class Config:
         if not self.api.mistral_api_key:
             validation_results["warnings"].append("Mistral API key not found - Mistral parser will be unavailable")
         
+        # Check RAG dependencies
+        if not self.api.openai_api_key:
+            validation_results["warnings"].append("OpenAI API key not found - RAG embeddings will be unavailable")
+        
+        if not self.api.google_api_key:
+            validation_results["warnings"].append("Google API key not found - RAG chat will be unavailable")
+        
         # Check tesseract setup
         if not self.ocr.tesseract_path and not os.path.exists("/usr/bin/tesseract"):
             validation_results["warnings"].append("Tesseract not found in system PATH - OCR functionality may be limited")
@@ -126,6 +190,14 @@ class Config:
             validation_results["errors"].append(f"Cannot create temp directory {self.app.temp_dir}: {e}")
             validation_results["valid"] = False
         
+        # Check RAG directories
+        try:
+            os.makedirs(self.rag.vector_store_path, exist_ok=True)
+            os.makedirs(self.rag.chat_history_path, exist_ok=True)
+        except Exception as e:
+            validation_results["errors"].append(f"Cannot create RAG directories: {e}")
+            validation_results["valid"] = False
+        
         return validation_results
     
     def get_available_parsers(self) -> list:

src/core/environment.py

@@ -154,6 +154,64 @@ class EnvironmentManager:
                 print(f"Error installing Docling: {e}")
                 return False
     
+    def check_rag_dependencies(self) -> bool:
+        """Check and install RAG dependencies (LangChain, ChromaDB, etc.)."""
+        rag_packages = [
+            ("langchain", "langchain>=0.3.0"),
+            ("langchain_openai", "langchain-openai>=0.2.0"),
+            ("langchain_google_genai", "langchain-google-genai>=2.0.0"),
+            ("langchain_chroma", "langchain-chroma>=0.1.0"),
+            ("langchain_text_splitters", "langchain-text-splitters>=0.3.0"),
+            ("chromadb", "chromadb>=0.5.0"),
+            ("sentence_transformers", "sentence-transformers>=3.0.0")
+        ]
+        
+        all_installed = True
+        
+        for import_name, install_name in rag_packages:
+            try:
+                if import_name == "langchain_openai":
+                    from langchain_openai import OpenAIEmbeddings
+                elif import_name == "langchain_google_genai":
+                    from langchain_google_genai import ChatGoogleGenerativeAI
+                elif import_name == "langchain_chroma":
+                    from langchain_chroma import Chroma
+                elif import_name == "langchain_text_splitters":
+                    from langchain_text_splitters import RecursiveCharacterTextSplitter
+                else:
+                    __import__(import_name)
+                
+                print(f"✅ {import_name} is installed")
+                
+            except ImportError:
+                print(f"WARNING: {import_name} not installed. Installing {install_name}...")
+                try:
+                    subprocess.run([sys.executable, "-m", "pip", "install", "-q", install_name], check=False)
+                    # Verify installation
+                    if import_name == "langchain_openai":
+                        from langchain_openai import OpenAIEmbeddings
+                    elif import_name == "langchain_google_genai":
+                        from langchain_google_genai import ChatGoogleGenerativeAI
+                    elif import_name == "langchain_chroma":
+                        from langchain_chroma import Chroma
+                    elif import_name == "langchain_text_splitters":
+                        from langchain_text_splitters import RecursiveCharacterTextSplitter
+                    else:
+                        __import__(import_name)
+                    
+                    print(f"✅ {import_name} installed successfully")
+                    
+                except (ImportError, Exception) as e:
+                    print(f"❌ Failed to install {import_name}: {e}")
+                    all_installed = False
+        
+        if all_installed:
+            print("✅ All RAG dependencies are available")
+        else:
+            print("⚠️ Some RAG dependencies failed to install - chat functionality may be limited")
+        
+        return all_installed
+    
     def load_environment_variables(self) -> bool:
         """Load environment variables from .env file."""
         try:
@@ -236,6 +294,7 @@ class EnvironmentManager:
         results["numpy"] = self.check_numpy()
         results["markitdown"] = self.check_markitdown()
         results["docling"] = self.check_docling()
+        results["rag_dependencies"] = self.check_rag_dependencies()
         
         # Load environment variables
         results["env_vars"] = self.load_environment_variables()

src/rag/__init__.py

@@ -0,0 +1,17 @@
+"""RAG (Retrieval-Augmented Generation) module for document chat functionality."""
+
+from .embeddings import embedding_manager
+from .chunking import document_chunker
+from .vector_store import vector_store_manager
+from .memory import chat_memory_manager
+from .chat_service import rag_chat_service
+from .ingestion import document_ingestion_service
+
+__all__ = [
+    "embedding_manager",
+    "document_chunker", 
+    "vector_store_manager",
+    "chat_memory_manager",
+    "rag_chat_service",
+    "document_ingestion_service"
+]

src/rag/chat_service.py

@@ -0,0 +1,367 @@
+"""RAG chat service with Gemini 2.5 Flash and streaming support."""
+
+import os
+import time
+from typing import List, Dict, Any, Optional, Generator, Tuple
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
+from langchain_core.documents import Document
+from src.rag.vector_store import vector_store_manager
+from src.rag.memory import chat_memory_manager
+from src.core.config import config
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+class ChatUsageLimiter:
+    """Manages chat usage limits to prevent abuse."""
+    
+    def __init__(self, max_messages_per_session: int = 50, max_messages_per_hour: int = 100):
+        """
+        Initialize usage limiter.
+        
+        Args:
+            max_messages_per_session: Maximum messages per chat session
+            max_messages_per_hour: Maximum messages per hour across all sessions
+        """
+        self.max_messages_per_session = max_messages_per_session
+        self.max_messages_per_hour = max_messages_per_hour
+        self.hourly_usage = {}  # Track usage by hour
+        
+        logger.info(f"Chat usage limiter initialized: {max_messages_per_session}/session, {max_messages_per_hour}/hour")
+    
+    def check_session_limit(self, session_message_count: int) -> Tuple[bool, str]:
+        """
+        Check if session has exceeded message limit.
+        
+        Args:
+            session_message_count: Number of messages in current session
+            
+        Returns:
+            Tuple of (allowed, reason_if_not_allowed)
+        """
+        if session_message_count >= self.max_messages_per_session:
+            return False, f"Session limit reached ({self.max_messages_per_session} messages per session). Please start a new chat."
+        return True, ""
+    
+    def check_hourly_limit(self) -> Tuple[bool, str]:
+        """
+        Check if hourly limit has been exceeded.
+        
+        Returns:
+            Tuple of (allowed, reason_if_not_allowed)
+        """
+        current_hour = int(time.time()) // 3600
+        
+        # Clean old entries (keep only last 2 hours)
+        hours_to_keep = [current_hour - 1, current_hour]
+        self.hourly_usage = {h: count for h, count in self.hourly_usage.items() if h in hours_to_keep}
+        
+        current_usage = self.hourly_usage.get(current_hour, 0)
+        
+        if current_usage >= self.max_messages_per_hour:
+            return False, f"Hourly limit reached ({self.max_messages_per_hour} messages per hour). Please try again later."
+        
+        return True, ""
+    
+    def record_usage(self) -> None:
+        """Record a message usage."""
+        current_hour = int(time.time()) // 3600
+        self.hourly_usage[current_hour] = self.hourly_usage.get(current_hour, 0) + 1
+    
+    def can_send_message(self, session_message_count: int) -> Tuple[bool, str]:
+        """
+        Check if user can send a message.
+        
+        Args:
+            session_message_count: Number of messages in current session
+            
+        Returns:
+            Tuple of (allowed, reason_if_not_allowed)
+        """
+        # Check session limit
+        session_ok, session_reason = self.check_session_limit(session_message_count)
+        if not session_ok:
+            return False, session_reason
+        
+        # Check hourly limit
+        hourly_ok, hourly_reason = self.check_hourly_limit()
+        if not hourly_ok:
+            return False, hourly_reason
+        
+        return True, ""
+
+class RAGChatService:
+    """RAG-powered chat service with document context."""
+    
+    def __init__(self):
+        """Initialize the RAG chat service."""
+        self.usage_limiter = ChatUsageLimiter(
+            max_messages_per_session=config.rag.max_messages_per_session,
+            max_messages_per_hour=config.rag.max_messages_per_hour
+        )
+        self._llm = None
+        self._rag_chain = None
+        
+        logger.info("RAG chat service initialized")
+    
+    def get_llm(self) -> ChatGoogleGenerativeAI:
+        """Get or create the Gemini LLM instance."""
+        if self._llm is None:
+            try:
+                google_api_key = config.api.google_api_key or os.getenv("GOOGLE_API_KEY")
+                
+                if not google_api_key:
+                    raise ValueError("Google API key not found. Please set GOOGLE_API_KEY in environment variables.")
+                
+                self._llm = ChatGoogleGenerativeAI(
+                    model="gemini-2.5-flash",  # Latest Gemini model
+                    google_api_key=google_api_key,
+                    temperature=0.1,
+                    max_tokens=4096,
+                    disable_streaming=False  # Enable streaming (new parameter name)
+                )
+                
+                logger.info("Gemini 2.5 Flash LLM initialized successfully")
+                
+            except Exception as e:
+                logger.error(f"Failed to initialize Gemini LLM: {e}")
+                raise
+        
+        return self._llm
+    
+    def create_rag_chain(self):
+        """Create the RAG chain for document-aware conversations."""
+        if self._rag_chain is None:
+            try:
+                llm = self.get_llm()
+                retriever = vector_store_manager.get_retriever(
+                    search_type="similarity",
+                    search_kwargs={"k": 4}
+                )
+                
+                # Create a prompt template for RAG
+                prompt_template = ChatPromptTemplate.from_template("""
+You are a helpful assistant that answers questions based on the provided document context.
+
+Instructions:
+1. Use the context provided to answer the user's question
+2. If the information is not in the context, say "I don't have enough information in the provided documents to answer that question"
+3. Always cite which parts of the documents you used for your answer
+4. Be concise but comprehensive
+5. If you find relevant tables or code blocks, include them in your response
+6. Maintain a conversational tone
+
+Context from documents:
+{context}
+
+Chat History:
+{chat_history}
+
+User Question: {question}
+""")
+                
+                def format_docs(docs: List[Document]) -> str:
+                    """Format retrieved documents for context."""
+                    if not docs:
+                        return "No relevant documents found."
+                    
+                    formatted = []
+                    for i, doc in enumerate(docs, 1):
+                        source = doc.metadata.get('source', 'Unknown')
+                        chunk_id = doc.metadata.get('chunk_id', f'chunk_{i}')
+                        
+                        formatted.append(f"Document {i} (Source: {source}, ID: {chunk_id}):\n{doc.page_content}")
+                    
+                    return "\n\n".join(formatted)
+                
+                def format_chat_history() -> str:
+                    """Format chat history for context."""
+                    history = chat_memory_manager.get_conversation_history(max_messages=10)
+                    if not history:
+                        return "No previous conversation."
+                    
+                    formatted = []
+                    for user_msg, assistant_msg in history[-5:]:  # Last 5 exchanges
+                        formatted.append(f"User: {user_msg}")
+                        formatted.append(f"Assistant: {assistant_msg}")
+                    
+                    return "\n".join(formatted)
+                
+                # Create the RAG chain
+                self._rag_chain = (
+                    {
+                        "context": retriever | format_docs,
+                        "chat_history": lambda _: format_chat_history(),
+                        "question": RunnablePassthrough()
+                    }
+                    | prompt_template
+                    | llm
+                    | StrOutputParser()
+                )
+                
+                logger.info("RAG chain created successfully")
+                
+            except Exception as e:
+                logger.error(f"Failed to create RAG chain: {e}")
+                raise
+    
+    def get_rag_chain(self):
+        """Get the RAG chain, creating it if necessary."""
+        if self._rag_chain is None:
+            self.create_rag_chain()
+        return self._rag_chain
+    
+    def chat_stream(self, user_message: str) -> Generator[str, None, None]:
+        """
+        Stream chat response using RAG.
+        
+        Args:
+            user_message: User's message
+            
+        Yields:
+            Chunks of the response as they're generated
+        """
+        try:
+            # Check usage limits
+            current_session = chat_memory_manager.current_session
+            session_message_count = len(current_session.messages) if current_session else 0
+            
+            can_send, reason = self.usage_limiter.can_send_message(session_message_count)
+            if not can_send:
+                yield f"❌ {reason}"
+                return
+            
+            # Record usage
+            self.usage_limiter.record_usage()
+            
+            # Add user message to memory
+            chat_memory_manager.add_message("user", user_message)
+            
+            # Get RAG chain
+            rag_chain = self.get_rag_chain()
+            
+            # Stream the response
+            response_chunks = []
+            for chunk in rag_chain.stream(user_message):
+                if chunk:
+                    response_chunks.append(chunk)
+                    yield chunk
+            
+            # Save complete response to memory
+            complete_response = "".join(response_chunks)
+            if complete_response.strip():
+                chat_memory_manager.add_message("assistant", complete_response)
+                
+                # Save session periodically
+                chat_memory_manager.save_session()
+            
+        except Exception as e:
+            error_msg = f"Error generating response: {str(e)}"
+            logger.error(error_msg)
+            yield f"❌ {error_msg}"
+    
+    def chat(self, user_message: str) -> str:
+        """
+        Get a complete chat response (non-streaming).
+        
+        Args:
+            user_message: User's message
+            
+        Returns:
+            Complete response string
+        """
+        try:
+            # Check usage limits
+            current_session = chat_memory_manager.current_session
+            session_message_count = len(current_session.messages) if current_session else 0
+            
+            can_send, reason = self.usage_limiter.can_send_message(session_message_count)
+            if not can_send:
+                return f"❌ {reason}"
+            
+            # Record usage
+            self.usage_limiter.record_usage()
+            
+            # Add user message to memory
+            chat_memory_manager.add_message("user", user_message)
+            
+            # Get RAG chain
+            rag_chain = self.get_rag_chain()
+            
+            # Get response
+            response = rag_chain.invoke(user_message)
+            
+            # Save response to memory
+            if response.strip():
+                chat_memory_manager.add_message("assistant", response)
+                chat_memory_manager.save_session()
+            
+            return response
+            
+        except Exception as e:
+            error_msg = f"Error generating response: {str(e)}"
+            logger.error(error_msg)
+            return f"❌ {error_msg}"
+    
+    def get_usage_stats(self) -> Dict[str, Any]:
+        """Get current usage statistics."""
+        current_session = chat_memory_manager.current_session
+        session_message_count = len(current_session.messages) if current_session else 0
+        
+        current_hour = int(time.time()) // 3600
+        hourly_count = self.usage_limiter.hourly_usage.get(current_hour, 0)
+        
+        return {
+            "session_messages": session_message_count,
+            "session_limit": self.usage_limiter.max_messages_per_session,
+            "hourly_messages": hourly_count,
+            "hourly_limit": self.usage_limiter.max_messages_per_hour,
+            "session_remaining": max(0, self.usage_limiter.max_messages_per_session - session_message_count),
+            "hourly_remaining": max(0, self.usage_limiter.max_messages_per_hour - hourly_count)
+        }
+    
+    def start_new_session(self, document_sources: Optional[List[str]] = None) -> str:
+        """Start a new chat session."""
+        session_id = chat_memory_manager.create_session(document_sources)
+        logger.info(f"Started new chat session: {session_id}")
+        return session_id
+    
+    def test_service(self) -> Dict[str, Any]:
+        """Test the RAG service components."""
+        results = {
+            "llm_available": False,
+            "vector_store_available": False,
+            "embeddings_available": False,
+            "errors": []
+        }
+        
+        try:
+            # Test LLM
+            llm = self.get_llm()
+            test_response = llm.invoke("Test message")
+            results["llm_available"] = True
+        except Exception as e:
+            results["errors"].append(f"LLM test failed: {str(e)}")
+        
+        try:
+            # Test vector store
+            vector_info = vector_store_manager.get_collection_info()
+            results["vector_store_available"] = "error" not in vector_info
+            results["document_count"] = vector_info.get("document_count", 0)
+        except Exception as e:
+            results["errors"].append(f"Vector store test failed: {str(e)}")
+        
+        try:
+            # Test embeddings
+            from src.rag.embeddings import embedding_manager
+            results["embeddings_available"] = embedding_manager.test_embedding_model()
+        except Exception as e:
+            results["errors"].append(f"Embeddings test failed: {str(e)}")
+        
+        return results
+
+# Global RAG chat service instance
+rag_chat_service = RAGChatService()

src/rag/chunking.py

@@ -0,0 +1,273 @@
+"""Text chunking strategies for RAG document processing."""
+
+import re
+from typing import List, Dict, Any, Tuple
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain_core.documents import Document
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+class MarkdownAwareChunker:
+    """Handles markdown-aware document chunking that preserves tables and structures."""
+    
+    def __init__(self, chunk_size: int = 1000, chunk_overlap: int = 200):
+        """
+        Initialize the markdown-aware document chunker.
+        
+        Args:
+            chunk_size: Maximum size of each chunk in characters
+            chunk_overlap: Number of characters to overlap between chunks
+        """
+        self.chunk_size = chunk_size
+        self.chunk_overlap = chunk_overlap
+        
+        # Initialize the text splitter with markdown-aware settings
+        self.text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=chunk_size,
+            chunk_overlap=chunk_overlap,
+            length_function=len,
+            separators=[
+                "\n\n",  # Paragraphs and sections
+                "\n# ",  # H1 headers
+                "\n## ", # H2 headers
+                "\n### ", # H3 headers
+                "\n\n---\n\n",  # Horizontal rules
+                "\n",    # Lines
+                " ",     # Words
+                ".",     # Sentences
+                ",",     # Clauses
+                ""       # Characters
+            ],
+            keep_separator=True,
+            add_start_index=True
+        )
+        
+        # Regex patterns for markdown structures
+        self.table_pattern = re.compile(
+            r'(\|.*\|.*\n)+(\|[-\s|:]+\|.*\n)(\|.*\|.*\n)*',
+            re.MULTILINE
+        )
+        
+        self.code_block_pattern = re.compile(
+            r'```[\s\S]*?```',
+            re.MULTILINE
+        )
+        
+        logger.info(f"Markdown-aware chunker initialized with chunk_size={chunk_size}, overlap={chunk_overlap}")
+    
+    def extract_markdown_structures(self, content: str) -> Tuple[List[Tuple[int, int, str]], str]:
+        """
+        Extract markdown tables and code blocks, replacing them with placeholders.
+        
+        Args:
+            content: Original markdown content
+            
+        Returns:
+            Tuple of (structures_list, content_with_placeholders)
+            where structures_list contains (start, end, type, content) tuples
+        """
+        structures = []
+        
+        # Find all tables
+        for match in self.table_pattern.finditer(content):
+            structures.append((
+                match.start(),
+                match.end(),
+                "table",
+                match.group()
+            ))
+        
+        # Find all code blocks
+        for match in self.code_block_pattern.finditer(content):
+            structures.append((
+                match.start(),
+                match.end(),
+                "code_block",
+                match.group()
+            ))
+        
+        # Sort by start position
+        structures.sort(key=lambda x: x[0])
+        
+        # Replace structures with placeholders
+        content_with_placeholders = content
+        offset = 0
+        
+        for i, (start, end, struct_type, struct_content) in enumerate(structures):
+            placeholder = f"\n\n__STRUCTURE_{i}_{struct_type.upper()}__\n\n"
+            
+            # Adjust positions based on previous replacements
+            adjusted_start = start - offset
+            adjusted_end = end - offset
+            
+            content_with_placeholders = (
+                content_with_placeholders[:adjusted_start] +
+                placeholder +
+                content_with_placeholders[adjusted_end:]
+            )
+            
+            # Update offset for next replacement
+            offset += (end - start) - len(placeholder)
+        
+        return structures, content_with_placeholders
+    
+    def restore_structures(self, chunks: List[str], structures: List[Tuple[int, int, str, str]]) -> List[str]:
+        """
+        Restore markdown structures in chunks, keeping tables and code blocks intact.
+        
+        Args:
+            chunks: List of text chunks with placeholders
+            structures: List of original structures
+            
+        Returns:
+            List of chunks with restored structures
+        """
+        restored_chunks = []
+        
+        for chunk in chunks:
+            restored_chunk = chunk
+            
+            # Find placeholders in this chunk
+            placeholder_pattern = re.compile(r'__STRUCTURE_(\d+)_(\w+)__')
+            
+            for match in placeholder_pattern.finditer(chunk):
+                structure_index = int(match.group(1))
+                
+                if structure_index < len(structures):
+                    original_structure = structures[structure_index][3]
+                    restored_chunk = restored_chunk.replace(match.group(), original_structure)
+            
+            restored_chunks.append(restored_chunk)
+        
+        return restored_chunks
+    
+    def chunk_document(self, content: str, source_metadata: Dict[str, Any]) -> List[Document]:
+        """
+        Chunk a markdown document while preserving tables and code blocks.
+        
+        Args:
+            content: The markdown content to chunk
+            source_metadata: Metadata about the source document
+            
+        Returns:
+            List of Document objects with chunked content and enhanced metadata
+        """
+        try:
+            # Extract markdown structures (tables, code blocks) and replace with placeholders
+            structures, content_with_placeholders = self.extract_markdown_structures(content)
+            
+            # Create a document object with placeholders
+            doc = Document(
+                page_content=content_with_placeholders,
+                metadata=source_metadata
+            )
+            
+            # Split the document into chunks
+            chunks = self.text_splitter.split_documents([doc])
+            
+            # Restore markdown structures in chunks
+            chunk_contents = [chunk.page_content for chunk in chunks]
+            restored_contents = self.restore_structures(chunk_contents, structures)
+            
+            # Create enhanced chunks with restored content
+            enhanced_chunks = []
+            for i, (chunk, restored_content) in enumerate(zip(chunks, restored_contents)):
+                # Add chunk-specific metadata
+                chunk.metadata.update({
+                    "chunk_index": i,
+                    "total_chunks": len(chunks),
+                    "chunk_size": len(restored_content),
+                    "chunk_id": f"{source_metadata.get('source_id', 'unknown')}_{i}",
+                    "has_table": "table" in restored_content.lower() and "|" in restored_content,
+                    "has_code": "```" in restored_content
+                })
+                
+                # Update the chunk content with restored structures
+                chunk.page_content = restored_content
+                enhanced_chunks.append(chunk)
+            
+            logger.info(f"Document chunked into {len(enhanced_chunks)} markdown-aware pieces")
+            return enhanced_chunks
+            
+        except Exception as e:
+            logger.error(f"Error chunking markdown document: {e}")
+            # Fallback to regular chunking if markdown processing fails
+            return self._fallback_chunk(content, source_metadata)
+    
+    def _fallback_chunk(self, content: str, source_metadata: Dict[str, Any]) -> List[Document]:
+        """Fallback chunking method if markdown-aware chunking fails."""
+        try:
+            doc = Document(page_content=content, metadata=source_metadata)
+            chunks = self.text_splitter.split_documents([doc])
+            
+            for i, chunk in enumerate(chunks):
+                chunk.metadata.update({
+                    "chunk_index": i,
+                    "total_chunks": len(chunks),
+                    "chunk_size": len(chunk.page_content),
+                    "chunk_id": f"{source_metadata.get('source_id', 'unknown')}_{i}"
+                })
+            
+            logger.warning(f"Used fallback chunking for {len(chunks)} pieces")
+            return chunks
+            
+        except Exception as e:
+            logger.error(f"Error in fallback chunking: {e}")
+            raise
+    
+    def chunk_multiple_documents(self, documents: List[Dict[str, Any]]) -> List[Document]:
+        """
+        Chunk multiple documents for batch processing.
+        
+        Args:
+            documents: List of dictionaries with 'content' and 'metadata' keys
+            
+        Returns:
+            List of chunked Document objects
+        """
+        all_chunks = []
+        
+        for doc_data in documents:
+            content = doc_data.get('content', '')
+            metadata = doc_data.get('metadata', {})
+            
+            if content.strip():  # Only process non-empty content
+                chunks = self.chunk_document(content, metadata)
+                all_chunks.extend(chunks)
+        
+        logger.info(f"Chunked {len(documents)} documents into {len(all_chunks)} total chunks")
+        return all_chunks
+    
+    def get_chunk_preview(self, chunks: List[Document], max_chunks: int = 5) -> str:
+        """
+        Generate a preview of chunks for debugging/logging.
+        
+        Args:
+            chunks: List of Document chunks
+            max_chunks: Maximum number of chunks to include in preview
+            
+        Returns:
+            String preview of chunks
+        """
+        preview = f"Document Chunks Preview ({len(chunks)} total chunks):\n"
+        preview += "=" * 50 + "\n"
+        
+        for i, chunk in enumerate(chunks[:max_chunks]):
+            has_table = chunk.metadata.get('has_table', False)
+            has_code = chunk.metadata.get('has_code', False)
+            
+            preview += f"Chunk {i + 1}:\n"
+            preview += f"  Length: {len(chunk.page_content)} characters\n"
+            preview += f"  Has Table: {has_table}, Has Code: {has_code}\n"
+            preview += f"  Metadata: {chunk.metadata}\n"
+            preview += f"  Content preview: {chunk.page_content[:100]}...\n"
+            preview += "-" * 30 + "\n"
+        
+        if len(chunks) > max_chunks:
+            preview += f"... and {len(chunks) - max_chunks} more chunks\n"
+        
+        return preview
+
+# Global chunker instance with optimized settings for markdown RAG
+document_chunker = MarkdownAwareChunker(chunk_size=1000, chunk_overlap=200)

src/rag/embeddings.py

@@ -0,0 +1,64 @@
+"""Embedding model management for RAG functionality."""
+
+import os
+from typing import Optional
+from langchain_openai import OpenAIEmbeddings
+from src.core.config import config
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+class EmbeddingManager:
+    """Manages embedding models for document vectorization."""
+    
+    def __init__(self):
+        self._embedding_model: Optional[OpenAIEmbeddings] = None
+        
+    def get_embedding_model(self) -> OpenAIEmbeddings:
+        """Get or create the OpenAI embedding model."""
+        if self._embedding_model is None:
+            try:
+                # Get OpenAI API key from config/environment
+                openai_api_key = config.api.openai_api_key or os.getenv("OPENAI_API_KEY")
+                
+                if not openai_api_key:
+                    raise ValueError("OpenAI API key not found. Please set OPENAI_API_KEY in environment variables.")
+                
+                self._embedding_model = OpenAIEmbeddings(
+                    model="text-embedding-3-small",
+                    openai_api_key=openai_api_key,
+                    chunk_size=1000,  # Process documents in chunks
+                    max_retries=3,
+                    timeout=30
+                )
+                
+                logger.info("OpenAI embedding model initialized successfully")
+                
+            except Exception as e:
+                logger.error(f"Failed to initialize OpenAI embedding model: {e}")
+                raise
+                
+        return self._embedding_model
+    
+    def test_embedding_model(self) -> bool:
+        """Test if the embedding model is working correctly."""
+        try:
+            embedding_model = self.get_embedding_model()
+            # Test with a simple text
+            test_text = "This is a test for embedding functionality."
+            embedding = embedding_model.embed_query(test_text)
+            
+            # Check if we got a valid embedding (list of floats)
+            if isinstance(embedding, list) and len(embedding) > 0 and isinstance(embedding[0], float):
+                logger.info("Embedding model test successful")
+                return True
+            else:
+                logger.error("Embedding model test failed: Invalid embedding format")
+                return False
+                
+        except Exception as e:
+            logger.error(f"Embedding model test failed: {e}")
+            return False
+
+# Global embedding manager instance
+embedding_manager = EmbeddingManager()

src/rag/ingestion.py

@@ -0,0 +1,304 @@
+"""Document ingestion pipeline for RAG functionality."""
+
+import os
+import hashlib
+from typing import List, Dict, Any, Optional, Tuple
+from pathlib import Path
+from datetime import datetime
+from langchain_core.documents import Document
+from src.rag.chunking import document_chunker
+from src.rag.vector_store import vector_store_manager
+from src.rag.embeddings import embedding_manager
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+class DocumentIngestionService:
+    """Service for ingesting documents into the RAG system."""
+    
+    def __init__(self):
+        """Initialize the document ingestion service."""
+        self.processed_documents = set()  # Track processed document hashes
+        logger.info("Document ingestion service initialized")
+    
+    def create_document_hash(self, content: str) -> str:
+        """Create a hash for document content to avoid duplicates."""
+        return hashlib.sha256(content.encode('utf-8')).hexdigest()[:16]
+    
+    def prepare_document_metadata(self, 
+                                source_path: Optional[str] = None, 
+                                doc_type: str = "markdown",
+                                additional_metadata: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
+        """
+        Prepare metadata for a document.
+        
+        Args:
+            source_path: Original document path
+            doc_type: Type of document (markdown, pdf, etc.)
+            additional_metadata: Additional metadata to include
+            
+        Returns:
+            Dictionary with document metadata
+        """
+        metadata = {
+            "source": source_path or "user_upload",
+            "doc_type": doc_type,
+            "processed_at": datetime.now().isoformat(),
+            "source_id": self.create_document_hash(source_path or ""),
+            "ingestion_version": "1.0"
+        }
+        
+        if additional_metadata:
+            metadata.update(additional_metadata)
+            
+        return metadata
+    
+    def ingest_markdown_content(self, 
+                              markdown_content: str, 
+                              source_path: Optional[str] = None,
+                              metadata: Optional[Dict[str, Any]] = None) -> Tuple[bool, str, Dict[str, Any]]:
+        """
+        Ingest markdown content into the RAG system.
+        
+        Args:
+            markdown_content: The markdown content to ingest
+            source_path: Optional source path/filename
+            metadata: Optional additional metadata
+            
+        Returns:
+            Tuple of (success, message, ingestion_stats)
+        """
+        try:
+            if not markdown_content or not markdown_content.strip():
+                return False, "No content provided for ingestion", {}
+            
+            # Create document hash to check for duplicates
+            content_hash = self.create_document_hash(markdown_content)
+            
+            if content_hash in self.processed_documents:
+                logger.info(f"Document already processed: {content_hash}")
+                return True, "Document already exists in the system", {"status": "duplicate"}
+            
+            # Prepare document metadata
+            doc_metadata = self.prepare_document_metadata(
+                source_path=source_path,
+                doc_type="markdown",
+                additional_metadata=metadata
+            )
+            doc_metadata["content_hash"] = content_hash
+            doc_metadata["content_length"] = len(markdown_content)
+            
+            # Chunk the document using markdown-aware chunking
+            logger.info(f"Chunking document: {content_hash}")
+            chunks = document_chunker.chunk_document(markdown_content, doc_metadata)
+            
+            if not chunks:
+                return False, "Failed to create document chunks", {}
+            
+            # Add chunks to vector store
+            logger.info(f"Adding {len(chunks)} chunks to vector store")
+            doc_ids = vector_store_manager.add_documents(chunks)
+            
+            if not doc_ids:
+                return False, "Failed to add documents to vector store", {}
+            
+            # Mark document as processed
+            self.processed_documents.add(content_hash)
+            
+            # Prepare ingestion statistics
+            ingestion_stats = {
+                "status": "success",
+                "content_hash": content_hash,
+                "total_chunks": len(chunks),
+                "document_ids": doc_ids,
+                "content_length": len(markdown_content),
+                "has_tables": any(chunk.metadata.get("has_table", False) for chunk in chunks),
+                "has_code": any(chunk.metadata.get("has_code", False) for chunk in chunks),
+                "processed_at": datetime.now().isoformat()
+            }
+            
+            success_msg = f"Successfully ingested document with {len(chunks)} chunks"
+            logger.info(f"{success_msg}: {content_hash}")
+            
+            return True, success_msg, ingestion_stats
+            
+        except Exception as e:
+            error_msg = f"Error during document ingestion: {str(e)}"
+            logger.error(error_msg)
+            return False, error_msg, {"status": "error", "error": str(e)}
+    
+    def ingest_from_conversion_result(self, conversion_result: Dict[str, Any]) -> Tuple[bool, str, Dict[str, Any]]:
+        """
+        Ingest a document from Markit conversion result.
+        
+        Args:
+            conversion_result: Dictionary containing conversion results from Markit
+            
+        Returns:
+            Tuple of (success, message, ingestion_stats)
+        """
+        try:
+            # Extract markdown content from conversion result
+            markdown_content = conversion_result.get("markdown_content", "")
+            
+            if not markdown_content:
+                return False, "No markdown content found in conversion result", {}
+            
+            # Extract metadata from conversion result
+            original_filename = conversion_result.get("original_filename", "unknown")
+            conversion_method = conversion_result.get("conversion_method", "unknown")
+            
+            additional_metadata = {
+                "original_filename": original_filename,
+                "conversion_method": conversion_method,
+                "file_size": conversion_result.get("file_size", 0),
+                "conversion_time": conversion_result.get("conversion_time", 0)
+            }
+            
+            # Ingest the markdown content
+            return self.ingest_markdown_content(
+                markdown_content=markdown_content,
+                source_path=original_filename,
+                metadata=additional_metadata
+            )
+            
+        except Exception as e:
+            error_msg = f"Error ingesting from conversion result: {str(e)}"
+            logger.error(error_msg)
+            return False, error_msg, {"status": "error", "error": str(e)}
+    
+    def get_ingestion_status(self) -> Dict[str, Any]:
+        """
+        Get current ingestion system status.
+        
+        Returns:
+            Dictionary with system status information
+        """
+        status = {
+            "processed_documents": len(self.processed_documents),
+            "embedding_model_available": False,
+            "vector_store_available": False,
+            "system_ready": False
+        }
+        
+        try:
+            # Check embedding model
+            status["embedding_model_available"] = embedding_manager.test_embedding_model()
+            
+            # Check vector store
+            collection_info = vector_store_manager.get_collection_info()
+            status["vector_store_available"] = "error" not in collection_info
+            status["total_documents_in_store"] = collection_info.get("document_count", 0)
+            
+            # System is ready if both components are available
+            status["system_ready"] = (
+                status["embedding_model_available"] and 
+                status["vector_store_available"]
+            )
+            
+        except Exception as e:
+            logger.error(f"Error checking ingestion status: {e}")
+            status["error"] = str(e)
+        
+        return status
+    
+    def clear_processed_documents(self) -> None:
+        """Clear the set of processed documents."""
+        self.processed_documents.clear()
+        logger.info("Cleared processed documents cache")
+    
+    def test_ingestion_pipeline(self) -> Dict[str, Any]:
+        """
+        Test the complete ingestion pipeline with sample content.
+        
+        Returns:
+            Dictionary with test results
+        """
+        test_results = {
+            "pipeline_test": False,
+            "chunking_test": False,
+            "embedding_test": False,
+            "vector_store_test": False,
+            "errors": []
+        }
+        
+        try:
+            # Test with sample markdown content
+            test_content = """# Test Document
+
+This is a test document for the RAG ingestion pipeline.
+
+## Features
+
+- Document chunking
+- Embedding generation
+- Vector store integration
+
+## Sample Table
+
+| Feature | Status | Priority |
+|---------|--------|----------|
+| Chunking | ✅ | High |
+| Embeddings | ✅ | High |
+| Vector Store | ✅ | Medium |
+
+```python
+# Sample code block
+def test_function():
+    return "Hello, RAG!"
+```
+
+This document contains various markdown elements to test the ingestion pipeline.
+"""
+            
+            # Test chunking
+            metadata = self.prepare_document_metadata(
+                source_path="test_document.md",
+                doc_type="markdown"
+            )
+            
+            chunks = document_chunker.chunk_document(test_content, metadata)
+            test_results["chunking_test"] = len(chunks) > 0
+            
+            if not test_results["chunking_test"]:
+                test_results["errors"].append("Chunking test failed: No chunks created")
+                return test_results
+            
+            # Test embedding
+            test_results["embedding_test"] = embedding_manager.test_embedding_model()
+            
+            if not test_results["embedding_test"]:
+                test_results["errors"].append("Embedding test failed")
+                return test_results
+            
+            # Test vector store (add and retrieve)
+            doc_ids = vector_store_manager.add_documents(chunks[:1])  # Test with one chunk
+            test_results["vector_store_test"] = len(doc_ids) > 0
+            
+            if test_results["vector_store_test"]:
+                # Test retrieval
+                search_results = vector_store_manager.similarity_search("test document", k=1)
+                test_results["vector_store_test"] = len(search_results) > 0
+            
+            if not test_results["vector_store_test"]:
+                test_results["errors"].append("Vector store test failed")
+                return test_results
+            
+            # Overall pipeline test
+            test_results["pipeline_test"] = (
+                test_results["chunking_test"] and 
+                test_results["embedding_test"] and 
+                test_results["vector_store_test"]
+            )
+            
+            logger.info(f"Ingestion pipeline test completed: {test_results['pipeline_test']}")
+            
+        except Exception as e:
+            error_msg = f"Pipeline test error: {str(e)}"
+            test_results["errors"].append(error_msg)
+            logger.error(error_msg)
+        
+        return test_results
+
+# Global document ingestion service instance
+document_ingestion_service = DocumentIngestionService()

src/rag/memory.py

@@ -0,0 +1,284 @@
+"""Chat history and memory management for RAG conversations."""
+
+import json
+import os
+from typing import List, Dict, Any, Optional, Tuple
+from datetime import datetime
+from pathlib import Path
+from dataclasses import dataclass, asdict
+from langchain_core.messages import BaseMessage, HumanMessage, AIMessage
+from src.core.config import config
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+@dataclass
+class ChatMessage:
+    """Represents a single chat message."""
+    role: str  # "user" or "assistant"
+    content: str
+    timestamp: str
+    sources: Optional[List[str]] = None  # Source documents used for the response
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return asdict(self)
+    
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> 'ChatMessage':
+        """Create from dictionary."""
+        return cls(**data)
+
+@dataclass
+class ChatSession:
+    """Represents a chat session with history."""
+    session_id: str
+    created_at: str
+    updated_at: str
+    messages: List[ChatMessage]
+    document_sources: List[str]  # Documents available in this session
+    
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert to dictionary."""
+        return {
+            "session_id": self.session_id,
+            "created_at": self.created_at,
+            "updated_at": self.updated_at,
+            "messages": [msg.to_dict() for msg in self.messages],
+            "document_sources": self.document_sources
+        }
+    
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> 'ChatSession':
+        """Create from dictionary."""
+        messages = [ChatMessage.from_dict(msg) for msg in data.get("messages", [])]
+        return cls(
+            session_id=data["session_id"],
+            created_at=data["created_at"],
+            updated_at=data["updated_at"],
+            messages=messages,
+            document_sources=data.get("document_sources", [])
+        )
+
+class ChatMemoryManager:
+    """Manages chat history and memory for RAG conversations."""
+    
+    def __init__(self, persist_directory: Optional[str] = None):
+        """
+        Initialize the chat memory manager.
+        
+        Args:
+            persist_directory: Directory to persist chat history
+        """
+        if persist_directory is None:
+            persist_directory = config.rag.chat_history_path
+        
+        self.persist_directory = Path(persist_directory)
+        self.persist_directory.mkdir(parents=True, exist_ok=True)
+        
+        self.current_session: Optional[ChatSession] = None
+        
+        logger.info(f"ChatMemoryManager initialized with persist_directory={self.persist_directory}")
+    
+    def create_session(self, document_sources: Optional[List[str]] = None) -> str:
+        """
+        Create a new chat session.
+        
+        Args:
+            document_sources: List of document sources available for this session
+            
+        Returns:
+            Session ID
+        """
+        session_id = f"session_{datetime.now().strftime('%Y%m%d_%H%M%S_%f')}"
+        now = datetime.now().isoformat()
+        
+        self.current_session = ChatSession(
+            session_id=session_id,
+            created_at=now,
+            updated_at=now,
+            messages=[],
+            document_sources=document_sources or []
+        )
+        
+        logger.info(f"Created new chat session: {session_id}")
+        return session_id
+    
+    def add_message(self, role: str, content: str, sources: Optional[List[str]] = None) -> None:
+        """
+        Add a message to the current session.
+        
+        Args:
+            role: "user" or "assistant"
+            content: Message content
+            sources: Source documents used (for assistant messages)
+        """
+        if self.current_session is None:
+            self.create_session()
+        
+        message = ChatMessage(
+            role=role,
+            content=content,
+            timestamp=datetime.now().isoformat(),
+            sources=sources
+        )
+        
+        self.current_session.messages.append(message)
+        self.current_session.updated_at = datetime.now().isoformat()
+        
+        logger.info(f"Added {role} message to session {self.current_session.session_id}")
+    
+    def get_conversation_history(self, max_messages: Optional[int] = None) -> List[Tuple[str, str]]:
+        """
+        Get conversation history in Gradio chat format.
+        
+        Args:
+            max_messages: Maximum number of messages to return
+            
+        Returns:
+            List of (user_message, assistant_message) tuples
+        """
+        if not self.current_session or not self.current_session.messages:
+            return []
+        
+        messages = self.current_session.messages
+        if max_messages:
+            messages = messages[-max_messages:]
+        
+        # Group messages into pairs
+        history = []
+        user_msg = None
+        
+        for msg in messages:
+            if msg.role == "user":
+                user_msg = msg.content
+            elif msg.role == "assistant" and user_msg is not None:
+                history.append((user_msg, msg.content))
+                user_msg = None
+        
+        return history
+    
+    def get_context_messages(self, max_context_length: int = 4000) -> List[BaseMessage]:
+        """
+        Get recent messages formatted for LangChain context.
+        
+        Args:
+            max_context_length: Maximum context length in characters
+            
+        Returns:
+            List of LangChain message objects
+        """
+        if not self.current_session or not self.current_session.messages:
+            return []
+        
+        context_messages = []
+        current_length = 0
+        
+        # Start from the most recent messages and work backwards
+        for msg in reversed(self.current_session.messages):
+            msg_length = len(msg.content)
+            
+            if current_length + msg_length > max_context_length:
+                break
+            
+            if msg.role == "user":
+                context_messages.insert(0, HumanMessage(content=msg.content))
+            elif msg.role == "assistant":
+                context_messages.insert(0, AIMessage(content=msg.content))
+            
+            current_length += msg_length
+        
+        logger.info(f"Retrieved {len(context_messages)} context messages ({current_length} chars)")
+        return context_messages
+    
+    def save_session(self) -> bool:
+        """
+        Save the current session to disk.
+        
+        Returns:
+            True if successful, False otherwise
+        """
+        if not self.current_session:
+            return False
+        
+        try:
+            session_file = self.persist_directory / f"{self.current_session.session_id}.json"
+            
+            with open(session_file, 'w', encoding='utf-8') as f:
+                json.dump(self.current_session.to_dict(), f, indent=2, ensure_ascii=False)
+            
+            logger.info(f"Saved session {self.current_session.session_id}")
+            return True
+            
+        except Exception as e:
+            logger.error(f"Error saving session: {e}")
+            return False
+    
+    def load_session(self, session_id: str) -> bool:
+        """
+        Load a session from disk.
+        
+        Args:
+            session_id: Session ID to load
+            
+        Returns:
+            True if successful, False otherwise
+        """
+        try:
+            session_file = self.persist_directory / f"{session_id}.json"
+            
+            if not session_file.exists():
+                logger.warning(f"Session file not found: {session_id}")
+                return False
+            
+            with open(session_file, 'r', encoding='utf-8') as f:
+                session_data = json.load(f)
+            
+            self.current_session = ChatSession.from_dict(session_data)
+            logger.info(f"Loaded session {session_id}")
+            return True
+            
+        except Exception as e:
+            logger.error(f"Error loading session {session_id}: {e}")
+            return False
+    
+    def list_sessions(self) -> List[Dict[str, Any]]:
+        """
+        List all saved sessions.
+        
+        Returns:
+            List of session metadata
+        """
+        sessions = []
+        
+        try:
+            for session_file in self.persist_directory.glob("session_*.json"):
+                try:
+                    with open(session_file, 'r', encoding='utf-8') as f:
+                        session_data = json.load(f)
+                    
+                    sessions.append({
+                        "session_id": session_data["session_id"],
+                        "created_at": session_data["created_at"],
+                        "updated_at": session_data["updated_at"],
+                        "message_count": len(session_data.get("messages", [])),
+                        "document_sources": session_data.get("document_sources", [])
+                    })
+                    
+                except Exception as e:
+                    logger.warning(f"Error reading session file {session_file}: {e}")
+                    
+        except Exception as e:
+            logger.error(f"Error listing sessions: {e}")
+        
+        # Sort by updated_at (most recent first)
+        sessions.sort(key=lambda x: x["updated_at"], reverse=True)
+        return sessions
+    
+    def clear_current_session(self) -> None:
+        """Clear the current session."""
+        self.current_session = None
+        logger.info("Cleared current session")
+
+# Global chat memory manager instance
+chat_memory_manager = ChatMemoryManager()

src/rag/vector_store.py

@@ -0,0 +1,230 @@
+"""Vector store management using Chroma for document storage and retrieval."""
+
+import os
+from typing import List, Optional, Dict, Any, Tuple
+from pathlib import Path
+from langchain_chroma import Chroma
+from langchain_core.documents import Document
+from langchain_core.vectorstores import VectorStoreRetriever
+from src.rag.embeddings import embedding_manager
+from src.core.config import config
+from src.core.logging_config import get_logger
+
+logger = get_logger(__name__)
+
+class VectorStoreManager:
+    """Manages Chroma vector store for document storage and retrieval."""
+    
+    def __init__(self, persist_directory: Optional[str] = None, collection_name: str = "markit_documents"):
+        """
+        Initialize the vector store manager.
+        
+        Args:
+            persist_directory: Directory to persist the vector database
+            collection_name: Name of the collection in the vector store
+        """
+        self.collection_name = collection_name
+        
+        # Set default persist directory
+        if persist_directory is None:
+            persist_directory = config.rag.vector_store_path
+        
+        self.persist_directory = str(Path(persist_directory).resolve())
+        
+        # Ensure the directory exists
+        os.makedirs(self.persist_directory, exist_ok=True)
+        
+        self._vector_store: Optional[Chroma] = None
+        
+        logger.info(f"VectorStoreManager initialized with persist_directory={self.persist_directory}")
+    
+    def get_vector_store(self) -> Chroma:
+        """Get or create the Chroma vector store."""
+        if self._vector_store is None:
+            try:
+                embedding_model = embedding_manager.get_embedding_model()
+                
+                self._vector_store = Chroma(
+                    collection_name=self.collection_name,
+                    embedding_function=embedding_model,
+                    persist_directory=self.persist_directory,
+                    collection_metadata={"hnsw:space": "cosine"}  # Use cosine similarity
+                )
+                
+                logger.info(f"Vector store initialized with collection '{self.collection_name}'")
+                
+            except Exception as e:
+                logger.error(f"Failed to initialize vector store: {e}")
+                raise
+        
+        return self._vector_store
+    
+    def add_documents(self, documents: List[Document]) -> List[str]:
+        """
+        Add documents to the vector store.
+        
+        Args:
+            documents: List of Document objects to add
+            
+        Returns:
+            List of document IDs that were added
+        """
+        try:
+            if not documents:
+                logger.warning("No documents provided to add to vector store")
+                return []
+            
+            vector_store = self.get_vector_store()
+            
+            # Generate unique IDs for documents
+            doc_ids = [f"doc_{i}_{hash(doc.page_content)}" for i, doc in enumerate(documents)]
+            
+            # Add documents to the vector store
+            added_ids = vector_store.add_documents(documents=documents, ids=doc_ids)
+            
+            logger.info(f"Added {len(added_ids)} documents to vector store")
+            return added_ids
+            
+        except Exception as e:
+            logger.error(f"Error adding documents to vector store: {e}")
+            raise
+    
+    def similarity_search(self, query: str, k: int = 4, score_threshold: Optional[float] = None) -> List[Document]:
+        """
+        Search for similar documents using semantic similarity.
+        
+        Args:
+            query: Search query
+            k: Number of documents to return
+            score_threshold: Minimum similarity score threshold
+            
+        Returns:
+            List of similar documents
+        """
+        try:
+            vector_store = self.get_vector_store()
+            
+            if score_threshold is not None:
+                # Use similarity search with score threshold
+                docs_with_scores = vector_store.similarity_search_with_relevance_scores(
+                    query=query,
+                    k=k,
+                    score_threshold=score_threshold
+                )
+                documents = [doc for doc, score in docs_with_scores]
+            else:
+                # Regular similarity search
+                documents = vector_store.similarity_search(query=query, k=k)
+            
+            logger.info(f"Found {len(documents)} similar documents for query: '{query[:50]}...'")
+            return documents
+            
+        except Exception as e:
+            logger.error(f"Error performing similarity search: {e}")
+            return []
+    
+    def get_retriever(self, search_type: str = "similarity", search_kwargs: Optional[Dict[str, Any]] = None) -> VectorStoreRetriever:
+        """
+        Get a retriever for the vector store.
+        
+        Args:
+            search_type: Type of search ("similarity", "mmr", "similarity_score_threshold")
+            search_kwargs: Additional search parameters
+            
+        Returns:
+            VectorStoreRetriever object
+        """
+        try:
+            vector_store = self.get_vector_store()
+            
+            if search_kwargs is None:
+                search_kwargs = {"k": 4}
+            
+            retriever = vector_store.as_retriever(
+                search_type=search_type,
+                search_kwargs=search_kwargs
+            )
+            
+            logger.info(f"Created retriever with search_type='{search_type}' and kwargs={search_kwargs}")
+            return retriever
+            
+        except Exception as e:
+            logger.error(f"Error creating retriever: {e}")
+            raise
+    
+    def get_collection_info(self) -> Dict[str, Any]:
+        """
+        Get information about the current collection.
+        
+        Returns:
+            Dictionary with collection information
+        """
+        try:
+            vector_store = self.get_vector_store()
+            
+            # Get collection count
+            count = vector_store._collection.count()
+            
+            info = {
+                "collection_name": self.collection_name,
+                "persist_directory": self.persist_directory,
+                "document_count": count,
+                "embedding_model": "text-embedding-3-small"
+            }
+            
+            logger.info(f"Collection info: {info}")
+            return info
+            
+        except Exception as e:
+            logger.error(f"Error getting collection info: {e}")
+            return {"error": str(e)}
+    
+    def delete_collection(self) -> bool:
+        """
+        Delete the current collection and reset the vector store.
+        
+        Returns:
+            True if successful, False otherwise
+        """
+        try:
+            if self._vector_store is not None:
+                self._vector_store.delete_collection()
+                self._vector_store = None
+                
+            logger.info(f"Deleted collection '{self.collection_name}'")
+            return True
+            
+        except Exception as e:
+            logger.error(f"Error deleting collection: {e}")
+            return False
+    
+    def search_with_metadata_filter(self, query: str, metadata_filter: Dict[str, Any], k: int = 4) -> List[Document]:
+        """
+        Search documents with metadata filtering.
+        
+        Args:
+            query: Search query
+            metadata_filter: Metadata filter conditions
+            k: Number of documents to return
+            
+        Returns:
+            List of filtered documents
+        """
+        try:
+            vector_store = self.get_vector_store()
+            
+            documents = vector_store.similarity_search(
+                query=query,
+                k=k,
+                filter=metadata_filter
+            )
+            
+            logger.info(f"Found {len(documents)} documents with metadata filter: {metadata_filter}")
+            return documents
+            
+        except Exception as e:
+            logger.error(f"Error searching with metadata filter: {e}")
+            return []
+
+# Global vector store manager instance
+vector_store_manager = VectorStoreManager()

src/ui/ui.py

@@ -14,6 +14,7 @@ from src.core.exceptions import (
     ConfigurationError
 )
 from src.core.logging_config import get_logger
+from src.rag import rag_chat_service, document_ingestion_service
 
 # Use centralized logging
 logger = get_logger(__name__)
@@ -166,11 +167,148 @@ def handle_convert(file_path, parser_name, ocr_method_name, output_format, is_ca
     html_output = f"<div class='output-container'>{formatted_content}</div>"
     
     logger.info("Conversion completed successfully")
+    
+    # Auto-ingest the converted document for RAG
+    try:
+        conversion_result = {
+            "markdown_content": content,
+            "original_filename": Path(file_path).name if file_path else "unknown",
+            "conversion_method": parser_name,
+            "file_size": Path(file_path).stat().st_size if file_path and Path(file_path).exists() else 0,
+            "conversion_time": 0  # Could be tracked if needed
+        }
+        
+        success, ingestion_msg, stats = document_ingestion_service.ingest_from_conversion_result(conversion_result)
+        if success:
+            logger.info(f"Document auto-ingested for RAG: {ingestion_msg}")
+        else:
+            logger.warning(f"Document ingestion failed: {ingestion_msg}")
+    except Exception as e:
+        logger.error(f"Error during auto-ingestion: {e}")
+    
     return html_output, download_file, gr.update(visible=False), gr.update(visible=True), gr.update(visible=False)
 
+def handle_chat_message(message, history):
+    """Handle a new chat message with streaming response."""
+    if not message or not message.strip():
+        return "", history
+    
+    try:
+        # Add user message to history
+        history = history or []
+        history.append({"role": "user", "content": message})
+        
+        # Add assistant message placeholder
+        history.append({"role": "assistant", "content": ""})
+        
+        # Get response from RAG service
+        response_text = ""
+        for chunk in rag_chat_service.chat_stream(message):
+            response_text += chunk
+            # Update the last message in history with the current response
+            history[-1]["content"] = response_text
+            yield "", history
+        
+        logger.info(f"Chat response completed for message: {message[:50]}...")
+        
+    except Exception as e:
+        error_msg = f"Error generating response: {str(e)}"
+        logger.error(error_msg)
+        if history and len(history) > 0:
+            history[-1]["content"] = f"❌ {error_msg}"
+        else:
+            history = [
+                {"role": "user", "content": message},
+                {"role": "assistant", "content": f"❌ {error_msg}"}
+            ]
+        yield "", history
+
+def start_new_chat_session():
+    """Start a new chat session."""
+    try:
+        session_id = rag_chat_service.start_new_session()
+        logger.info(f"Started new chat session: {session_id}")
+        return [], f"✅ New chat session started: {session_id}"
+    except Exception as e:
+        error_msg = f"Error starting new session: {str(e)}"
+        logger.error(error_msg)
+        return [], f"❌ {error_msg}"
+
+def get_chat_status():
+    """Get current chat system status."""
+    try:
+        # Check ingestion status
+        ingestion_status = document_ingestion_service.get_ingestion_status()
+        
+        # Check usage stats
+        usage_stats = rag_chat_service.get_usage_stats()
+        
+        # Modern status card design with better styling
+        status_html = f"""
+        <div class="status-card">
+            <div class="status-header">
+                <h3>💬 Chat System Status</h3>
+                <div class="status-indicator {'status-ready' if ingestion_status.get('system_ready', False) else 'status-not-ready'}">
+                    {'🟢 READY' if ingestion_status.get('system_ready', False) else '🔴 NOT READY'}
+                </div>
+            </div>
+            
+            <div class="status-grid">
+                <div class="status-item">
+                    <div class="status-label">Documents Processed</div>
+                    <div class="status-value">{ingestion_status.get('processed_documents', 0)}</div>
+                </div>
+                <div class="status-item">
+                    <div class="status-label">Vector Store</div>
+                    <div class="status-value">{ingestion_status.get('total_documents_in_store', 0)} docs</div>
+                </div>
+                <div class="status-item">
+                    <div class="status-label">Session Usage</div>
+                    <div class="status-value">{usage_stats.get('session_messages', 0)}/{usage_stats.get('session_limit', 50)}</div>
+                </div>
+                <div class="status-item">
+                    <div class="status-label">Hourly Usage</div>
+                    <div class="status-value">{usage_stats.get('hourly_messages', 0)}/{usage_stats.get('hourly_limit', 100)}</div>
+                </div>
+            </div>
+            
+            <div class="status-services">
+                <div class="service-status {'service-ready' if ingestion_status.get('embedding_model_available', False) else 'service-error'}">
+                    <span class="service-icon">🧠</span>
+                    <span>Embedding Model</span>
+                    <span class="service-indicator">{'✅' if ingestion_status.get('embedding_model_available', False) else '❌'}</span>
+                </div>
+                <div class="service-status {'service-ready' if ingestion_status.get('vector_store_available', False) else 'service-error'}">
+                    <span class="service-icon">🗄️</span>
+                    <span>Vector Store</span>
+                    <span class="service-indicator">{'✅' if ingestion_status.get('vector_store_available', False) else '❌'}</span>
+                </div>
+            </div>
+        </div>
+        """
+        
+        return status_html
+        
+    except Exception as e:
+        error_msg = f"Error getting chat status: {str(e)}"
+        logger.error(error_msg)
+        return f"""
+        <div class="status-card status-error">
+            <div class="status-header">
+                <h3>❌ System Error</h3>
+            </div>
+            <p class="error-message">{error_msg}</p>
+        </div>
+        """
+
 def create_ui():
     with gr.Blocks(css="""
-        /* Simple output container with only one scrollbar */
+        /* Global styles */
+        .gradio-container {
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+        }
+        
+        /* Document converter styles */
         .output-container {
             max-height: 420px;
             overflow-y: auto;
@@ -178,7 +316,6 @@ def create_ui():
             padding: 10px;
         }
         
-        /* Hide any scrollbars from parent containers */
         .gradio-container .prose {
             overflow: visible;
         }
@@ -190,115 +327,532 @@ def create_ui():
             margin-top: 10px; 
         }
         
-        /* Add margin above the provider/OCR options row */
         .provider-options-row {
             margin-top: 15px;
             margin-bottom: 15px;
         }
+        
+        /* Chat Tab Styles - Complete redesign */
+        .chat-tab-container {
+            max-width: 1200px;
+            margin: 0 auto;
+            padding: 20px;
+        }
+        
+        .chat-header {
+            text-align: center;
+            margin-bottom: 30px;
+            padding: 20px;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            border-radius: 15px;
+            color: white;
+            box-shadow: 0 4px 15px rgba(0,0,0,0.1);
+        }
+        
+        .chat-header h2 {
+            margin: 0;
+            font-size: 1.8em;
+            font-weight: 600;
+        }
+        
+        .chat-header p {
+            margin: 10px 0 0 0;
+            opacity: 0.9;
+            font-size: 1.1em;
+        }
+        
+        /* Status Card Styling */
+        .status-card {
+            background: #ffffff;
+            border: 1px solid #e1e5e9;
+            border-radius: 12px;
+            padding: 20px;
+            margin-bottom: 25px;
+            box-shadow: 0 2px 10px rgba(0,0,0,0.05);
+            transition: all 0.3s ease;
+        }
+        
+        .status-card:hover {
+            box-shadow: 0 4px 20px rgba(0,0,0,0.1);
+        }
+        
+        .status-header {
+            display: flex;
+            justify-content: space-between;
+            align-items: center;
+            margin-bottom: 20px;
+            padding-bottom: 15px;
+            border-bottom: 2px solid #f0f2f5;
+        }
+        
+        .status-header h3 {
+            margin: 0;
+            color: #2c3e50;
+            font-size: 1.3em;
+            font-weight: 600;
+        }
+        
+        .status-indicator {
+            padding: 8px 16px;
+            border-radius: 25px;
+            font-weight: 600;
+            font-size: 0.9em;
+            letter-spacing: 0.5px;
+        }
+        
+        .status-ready {
+            background: #d4edda;
+            color: #155724;
+            border: 1px solid #c3e6cb;
+        }
+        
+        .status-not-ready {
+            background: #f8d7da;
+            color: #721c24;
+            border: 1px solid #f5c6cb;
+        }
+        
+        .status-grid {
+            display: grid;
+            grid-template-columns: repeat(auto-fit, minmax(150px, 1fr));
+            gap: 15px;
+            margin-bottom: 20px;
+        }
+        
+        .status-item {
+            background: #f8f9fa;
+            padding: 15px;
+            border-radius: 8px;
+            text-align: center;
+            border: 1px solid #e9ecef;
+        }
+        
+        .status-label {
+            font-size: 0.85em;
+            color: #6c757d;
+            margin-bottom: 5px;
+            font-weight: 500;
+        }
+        
+        .status-value {
+            font-size: 1.4em;
+            font-weight: 700;
+            color: #495057;
+        }
+        
+        .status-services {
+            display: flex;
+            gap: 15px;
+            flex-wrap: wrap;
+        }
+        
+        .service-status {
+            display: flex;
+            align-items: center;
+            gap: 8px;
+            padding: 10px 15px;
+            border-radius: 8px;
+            font-weight: 500;
+            flex: 1;
+            min-width: 200px;
+        }
+        
+        .service-ready {
+            background: #d4edda;
+            color: #155724;
+            border: 1px solid #c3e6cb;
+        }
+        
+        .service-error {
+            background: #f8d7da;
+            color: #721c24;
+            border: 1px solid #f5c6cb;
+        }
+        
+        .service-icon {
+            font-size: 1.2em;
+        }
+        
+        .service-indicator {
+            margin-left: auto;
+        }
+        
+        .status-error {
+            border-color: #dc3545;
+            background: #f8d7da;
+        }
+        
+        .error-message {
+            color: #721c24;
+            margin: 0;
+            font-weight: 500;
+        }
+        
+        /* Control buttons styling */
+        .control-buttons {
+            display: flex;
+            gap: 12px;
+            justify-content: flex-end;
+            margin-bottom: 25px;
+        }
+        
+        .control-btn {
+            padding: 10px 20px;
+            border-radius: 8px;
+            font-weight: 500;
+            transition: all 0.3s ease;
+            border: none;
+            cursor: pointer;
+        }
+        
+        .btn-refresh {
+            background: #17a2b8;
+            color: white;
+        }
+        
+        .btn-refresh:hover {
+            background: #138496;
+            transform: translateY(-1px);
+        }
+        
+        .btn-new-session {
+            background: #28a745;
+            color: white;
+        }
+        
+        .btn-new-session:hover {
+            background: #218838;
+            transform: translateY(-1px);
+        }
+        
+        /* Chat interface styling */
+        .chat-main-container {
+            background: #ffffff;
+            border-radius: 15px;
+            box-shadow: 0 4px 20px rgba(0,0,0,0.08);
+            overflow: hidden;
+            margin-bottom: 25px;
+        }
+        
+        .chat-container {
+            background: #ffffff;
+            border-radius: 12px;
+            border: 1px solid #e1e5e9;
+            overflow: hidden;
+        }
+        
+        /* Custom chatbot styling */
+        .gradio-chatbot {
+            border: none !important;
+            background: #ffffff;
+        }
+        
+        .gradio-chatbot .message {
+            padding: 15px 20px;
+            margin: 10px;
+            border-radius: 12px;
+        }
+        
+        .gradio-chatbot .message.user {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            margin-left: 50px;
+        }
+        
+        .gradio-chatbot .message.assistant {
+            background: #f8f9fa;
+            border: 1px solid #e9ecef;
+            margin-right: 50px;
+        }
+        
+        /* Input area styling */
+        .chat-input-container {
+            background: #ffffff;
+            padding: 20px;
+            border-top: 1px solid #e1e5e9;
+            border-radius: 0 0 15px 15px;
+        }
+        
+        .input-row {
+            display: flex;
+            gap: 12px;
+            align-items: center;
+        }
+        
+        .message-input {
+            flex: 1;
+            border: 2px solid #e1e5e9;
+            border-radius: 25px;
+            padding: 12px 20px;
+            font-size: 1em;
+            transition: all 0.3s ease;
+            resize: none;
+            max-height: 120px;
+            min-height: 48px;
+        }
+        
+        .message-input:focus {
+            border-color: #667eea;
+            box-shadow: 0 0 0 3px rgba(102, 126, 234, 0.1);
+            outline: none;
+        }
+        
+        .send-button {
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            color: white;
+            border: none;
+            border-radius: 12px;
+            padding: 12px 24px;
+            min-width: 80px;
+            height: 48px;
+            margin-right: 10px;
+            cursor: pointer;
+            transition: all 0.3s ease;
+            display: flex;
+            align-items: center;
+            justify-content: center;
+            font-size: 1em;
+            font-weight: 600;
+            letter-spacing: 0.5px;
+        }
+        
+        .send-button:hover {
+            transform: scale(1.05);
+            box-shadow: 0 4px 15px rgba(102, 126, 234, 0.3);
+        }
+        
+        /* Session info styling */
+        .session-info {
+            background: #e7f3ff;
+            border: 1px solid #b3d9ff;
+            border-radius: 8px;
+            padding: 15px;
+            color: #0056b3;
+            font-weight: 500;
+            text-align: center;
+        }
+        
+        /* Responsive design */
+        @media (max-width: 768px) {
+            .chat-tab-container {
+                padding: 10px;
+            }
+            
+            .status-grid {
+                grid-template-columns: repeat(2, 1fr);
+            }
+            
+            .service-status {
+                min-width: 100%;
+            }
+            
+            .control-buttons {
+                flex-direction: column;
+                gap: 8px;
+            }
+            
+            .gradio-chatbot .message.user {
+                margin-left: 20px;
+            }
+            
+            .gradio-chatbot .message.assistant {
+                margin-right: 20px;
+            }
+        }
     """) as demo:
-        # Simple title - no fancy HTML or CSS
-        gr.Markdown("## Markit: Document to Markdown Converter")
-        
-        # State to track if cancellation is requested
-        cancel_requested = gr.State(False)
-        # State to store the conversion thread
-        conversion_thread = gr.State(None)
-        # State to store the output format (fixed to Markdown)
-        output_format_state = gr.State("Markdown")
-
-        # File input first
-        file_input = gr.File(label="Upload Document", type="filepath")
+        # Modern title with better styling
+        gr.Markdown("""
+        # 🚀 Markit
+        ## Document to Markdown Converter with RAG Chat
+        """)
         
-        # Provider and OCR options below the file input
-        with gr.Row(elem_classes=["provider-options-row"]):
-            with gr.Column(scale=1):
-                parser_names = ParserRegistry.get_parser_names()
+        with gr.Tabs():
+            # Document Converter Tab
+            with gr.TabItem("📄 Document Converter"):
+                # State to track if cancellation is requested
+                cancel_requested = gr.State(False)
+                # State to store the conversion thread
+                conversion_thread = gr.State(None)
+                # State to store the output format (fixed to Markdown)
+                output_format_state = gr.State("Markdown")
+
+                # File input first
+                file_input = gr.File(label="Upload Document", type="filepath")
                 
-                # Make MarkItDown the default parser if available
-                default_parser = next((p for p in parser_names if p == "MarkItDown"), parser_names[0] if parser_names else "PyPdfium")
+                # Provider and OCR options below the file input
+                with gr.Row(elem_classes=["provider-options-row"]):
+                    with gr.Column(scale=1):
+                        parser_names = ParserRegistry.get_parser_names()
+                        
+                        # Make MarkItDown the default parser if available
+                        default_parser = next((p for p in parser_names if p == "MarkItDown"), parser_names[0] if parser_names else "PyPdfium")
+                        
+                        provider_dropdown = gr.Dropdown(
+                            label="Provider",
+                            choices=parser_names,
+                            value=default_parser,
+                            interactive=True
+                        )
+                    with gr.Column(scale=1):
+                        default_ocr_options = ParserRegistry.get_ocr_options(default_parser)
+                        default_ocr = default_ocr_options[0] if default_ocr_options else "No OCR"
+                        
+                        ocr_dropdown = gr.Dropdown(
+                            label="OCR Options",
+                            choices=default_ocr_options,
+                            value=default_ocr,
+                            interactive=True
+                        )
                 
-                provider_dropdown = gr.Dropdown(
-                    label="Provider",
-                    choices=parser_names,
-                    value=default_parser,
-                    interactive=True
+                # Simple output container with just one scrollbar
+                file_display = gr.HTML(
+                    value="<div class='output-container'></div>",
+                    label="Converted Content"
                 )
-            with gr.Column(scale=1):
-                default_ocr_options = ParserRegistry.get_ocr_options(default_parser)
-                default_ocr = default_ocr_options[0] if default_ocr_options else "No OCR"
                 
-                ocr_dropdown = gr.Dropdown(
-                    label="OCR Options",
-                    choices=default_ocr_options,
-                    value=default_ocr,
-                    interactive=True
+                file_download = gr.File(label="Download File")
+                
+                # Processing controls row
+                with gr.Row(elem_classes=["processing-controls"]):
+                    convert_button = gr.Button("Convert", variant="primary")
+                    cancel_button = gr.Button("Cancel", variant="stop", visible=False)
+
+                # Event handlers for document converter
+                provider_dropdown.change(
+                    lambda p: gr.Dropdown(
+                        choices=["Plain Text", "Formatted Text"] if "GOT-OCR" in p else ParserRegistry.get_ocr_options(p),
+                        value="Plain Text" if "GOT-OCR" in p else (ParserRegistry.get_ocr_options(p)[0] if ParserRegistry.get_ocr_options(p) else None)
+                    ),
+                    inputs=[provider_dropdown],
+                    outputs=[ocr_dropdown]
                 )
-        
-        # Simple output container with just one scrollbar
-        file_display = gr.HTML(
-            value="<div class='output-container'></div>",
-            label="Converted Content"
-        )
-        
-        file_download = gr.File(label="Download File")
-        
-        # Processing controls row
-        with gr.Row(elem_classes=["processing-controls"]):
-            convert_button = gr.Button("Convert", variant="primary")
-            cancel_button = gr.Button("Cancel", variant="stop", visible=False)
 
-        # Event handlers
-        provider_dropdown.change(
-            lambda p: gr.Dropdown(
-                choices=["Plain Text", "Formatted Text"] if "GOT-OCR" in p else ParserRegistry.get_ocr_options(p),
-                value="Plain Text" if "GOT-OCR" in p else (ParserRegistry.get_ocr_options(p)[0] if ParserRegistry.get_ocr_options(p) else None)
-            ),
-            inputs=[provider_dropdown],
-            outputs=[ocr_dropdown]
-        )
+                # Reset cancel flag when starting conversion
+                def start_conversion():
+                    global conversion_cancelled
+                    conversion_cancelled.clear()
+                    logger.info("Starting conversion with cancellation flag cleared")
+                    return gr.update(visible=False), gr.update(visible=True), False
 
-        # Reset cancel flag when starting conversion
-        def start_conversion():
-            global conversion_cancelled
-            conversion_cancelled.clear()
-            logger.info("Starting conversion with cancellation flag cleared")
-            return gr.update(visible=False), gr.update(visible=True), False
+                # Set cancel flag and terminate thread when cancel button is clicked
+                def request_cancellation(thread):
+                    global conversion_cancelled
+                    conversion_cancelled.set()
+                    logger.info("Cancel button clicked, cancellation flag set")
+                    
+                    # Try to join the thread with a timeout
+                    if thread is not None:
+                        logger.info(f"Attempting to join conversion thread: {thread}")
+                        thread.join(timeout=0.5)
+                        if thread.is_alive():
+                            logger.warning("Thread did not finish within timeout")
+                    
+                    # Add immediate feedback to the user
+                    return gr.update(visible=True), gr.update(visible=False), True, None
 
-        # Set cancel flag and terminate thread when cancel button is clicked
-        def request_cancellation(thread):
-            global conversion_cancelled
-            conversion_cancelled.set()
-            logger.info("Cancel button clicked, cancellation flag set")
-            
-            # Try to join the thread with a timeout
-            if thread is not None:
-                logger.info(f"Attempting to join conversion thread: {thread}")
-                thread.join(timeout=0.5)
-                if thread.is_alive():
-                    logger.warning("Thread did not finish within timeout")
-            
-            # Add immediate feedback to the user
-            return gr.update(visible=True), gr.update(visible=False), True, None
+                # Start conversion sequence
+                convert_button.click(
+                    fn=start_conversion,
+                    inputs=[],
+                    outputs=[convert_button, cancel_button, cancel_requested],
+                    queue=False  # Execute immediately
+                ).then(
+                    fn=handle_convert,
+                    inputs=[file_input, provider_dropdown, ocr_dropdown, output_format_state, cancel_requested],
+                    outputs=[file_display, file_download, convert_button, cancel_button, conversion_thread]
+                )
+                
+                # Handle cancel button click
+                cancel_button.click(
+                    fn=request_cancellation,
+                    inputs=[conversion_thread],
+                    outputs=[convert_button, cancel_button, cancel_requested, conversion_thread],
+                    queue=False  # Execute immediately
+                )
 
-        # Start conversion sequence
-        convert_button.click(
-            fn=start_conversion,
-            inputs=[],
-            outputs=[convert_button, cancel_button, cancel_requested],
-            queue=False  # Execute immediately
-        ).then(
-            fn=handle_convert,
-            inputs=[file_input, provider_dropdown, ocr_dropdown, output_format_state, cancel_requested],
-            outputs=[file_display, file_download, convert_button, cancel_button, conversion_thread]
-        )
-        
-        # Handle cancel button click
-        cancel_button.click(
-            fn=request_cancellation,
-            inputs=[conversion_thread],
-            outputs=[convert_button, cancel_button, cancel_requested, conversion_thread],
-            queue=False  # Execute immediately
-        )
+            # Chat Tab - Completely redesigned
+            with gr.TabItem("💬 Chat with Documents"):
+                with gr.Column(elem_classes=["chat-tab-container"]):
+                    # Modern header
+                    gr.HTML("""
+                    <div class="chat-header">
+                        <h2>💬 Chat with your converted documents</h2>
+                        <p>Ask questions about your documents using advanced RAG technology</p>
+                    </div>
+                    """)
+                    
+                    # Status section with modern design
+                    status_display = gr.HTML(value=get_chat_status())
+                    
+                    # Control buttons
+                    with gr.Row(elem_classes=["control-buttons"]):
+                        refresh_status_btn = gr.Button("🔄 Refresh Status", elem_classes=["control-btn", "btn-refresh"])
+                        new_session_btn = gr.Button("🆕 New Session", elem_classes=["control-btn", "btn-new-session"])
+                    
+                    # Main chat interface
+                    with gr.Column(elem_classes=["chat-main-container"]):
+                        chatbot = gr.Chatbot(
+                            elem_classes=["chat-container"],
+                            height=500,
+                            show_label=False,
+                            show_share_button=False,
+                            bubble_full_width=False,
+                            type="messages",
+                            placeholder="Start a conversation by asking questions about your documents..."
+                        )
+                        
+                        # Input area
+                        with gr.Row(elem_classes=["input-row"]):
+                            msg_input = gr.Textbox(
+                                placeholder="Ask questions about your documents...",
+                                show_label=False,
+                                scale=5,
+                                lines=1,
+                                max_lines=3,
+                                elem_classes=["message-input"]
+                            )
+                            send_btn = gr.Button("Submit", elem_classes=["send-button"], scale=0)
+                    
+                    # Session info with better styling
+                    session_info = gr.HTML(
+                        value='<div class="session-info">No active session - Click "New Session" to start</div>'
+                    )
+                
+                # Event handlers for chat
+                def clear_input():
+                    return ""
+                
+                # Send message when button clicked or Enter pressed
+                msg_input.submit(
+                    fn=handle_chat_message,
+                    inputs=[msg_input, chatbot],
+                    outputs=[msg_input, chatbot]
+                )
+                
+                send_btn.click(
+                    fn=handle_chat_message,
+                    inputs=[msg_input, chatbot],
+                    outputs=[msg_input, chatbot]
+                )
+                
+                # New session handler with improved feedback
+                def enhanced_new_session():
+                    history, info = start_new_chat_session()
+                    session_html = f'<div class="session-info">{info}</div>'
+                    return history, session_html
+                
+                new_session_btn.click(
+                    fn=enhanced_new_session,
+                    inputs=[],
+                    outputs=[chatbot, session_info]
+                )
+                
+                # Refresh status handler
+                refresh_status_btn.click(
+                    fn=get_chat_status,
+                    inputs=[],
+                    outputs=[status_display]
+                )
 
     return demo