Documentation and cleaning the code

.gitignore

@@ -12,4 +12,8 @@ __pycache__/
 .cache/
 .DS_Store
 output*
-storage/*.json
+<<<<<<< HEAD
+*.csv
+=======
+storage/*.json
+>>>>>>> 8d902546c2f7d4238e340aadff3f1a20bfb4fee7

README.md

@@ -1 +1,133 @@
-# agentic-chatbot
+# 🤖 Agentic Chatbot: Navigating Red Hat Internal resources from THE SOURCE
+
+## Overview
+
+The Agentic Chatbot is an intelligent assistant designed to help Red Hat associates navigate internal resources efficiently.
+
+## Project Goals
+
+- Simplify access to internal documentation
+- Reduce onboarding friction
+- Provide intelligent, context-aware resource discovery
+
+## Use Case
+
+Instead of navigating the Source portal manually, associates can ask:
+
+- "What is the process for updating a Red Hat product?"
+- "Where can I find the latest documentation for a specific product?"
+- "How do I resolve a common issue with a Red Hat product?"
+
+And receive:
+
+- Direct answers
+- Links to the exact documents/pages
+- Context-aware follow-up suggestions
+
+## Technical Architecture
+
+### Core Components
+
+| Layer         | Technology                          |
+|---------------|--------------------------------------|
+| **Backend**   | Python + FastAPI / Flask             |
+| **LLMs**      | OpenAI GPT, Ollama, Hugging Face     |
+| **Vector DB** | ChromaDB / FAISS                     |
+| **Embeddings**| Hugging Face Transformers            |
+| **Frontend**  | Slack Bot (via Slack Bolt SDK)|
+
+### Agentic Workflow
+
+The chatbot operates on a **multi-agent architecture**, where each agent specializes in a sub-task:
+
+- **Query Understanding Agent**
+  Interprets user inputs into actionable formats.
+
+- **Content Indexing Agent**
+  Parses documents and pushes them to a vector database after generating embeddings.
+
+- **Response Generation Agent**
+  Retrieves relevant data and composes human-like, informative replies.
+
+## Slack Integration
+
+The chatbot is tightly integrated with Slack for enterprise accessibility:
+
+- Users interact via DM or threads
+- Replies include embedded links to Source documents
+- Backend uses Slack Bolt SDK with FastAPI
+- Events are securely handled with workspace-level tokens
+
+## Key Features
+
+- **Semantic Search**
+  Leverages vector embeddings to match intent with content, even for vague queries
+
+- **Model Flexibility**
+  Easily swap between OpenAI, Hugging Face, or locally hosted models via Ollama
+
+- **Microservice Architecture**
+  Modular design allows for scaling and independent agent upgrades
+
+- **Multilingual Support**
+  Via Hugging Face embeddings and tokenizers
+
+- **Secure Document Handling**
+  Respects access controls and data privacy protocols
+
+## Impact
+
+- **Significantly reduces time** spent navigating documentation manually
+- **Improves onboarding** experience for new Red Hatters
+- **Promotes self-service** culture and reduces dependency on internal channels
+- **Lays foundation** for enterprise-grade knowledge retrieval systems
+
+## Future Scope
+
+- **EagleView API Integration**
+  Post-authentication, the chatbot can dynamically fetch the complete Source portal data and continuously update its knowledge base
+
+- **Scalable to Entire Source Ecosystem**
+  Once EagleView API access is enabled, the chatbot will be able to answer **all** questions across teams and departments
+
+- **Credential-Aware Access Control**
+  Role-based response customization based on user credentials
+
+- **Intelligent Logging & Feedback Loop**
+  Enable query analytics to improve answers through fine-tuning
+
+- **Horizontally Scalable Architecture**
+  Multi-agent system allows parallel processing, independent agent upgrades, and multi-tenant support
+
+## Installation
+
+```bash
+# Clone the repository
+git clone https://github.com/your-org/agentic-chatbot.git
+
+# Install dependencies
+pip install -r requirements.txt
+
+# Set up environment variables
+cp sample.env .env
+# Edit .env with your configurations
+```
+
+## Quick Start
+
+### Running the Slack Bot
+```bash
+python services/slack_service.py
+```
+
+## Contributing
+
+1. Fork the repository
+2. Create your feature branch (`git checkout -b feature/AmazingFeature`)
+3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
+4. Push to the branch (`git push origin feature/AmazingFeature`)
+5. Open a Pull Request
+
+## License
+
+Distributed under the MIT License. See `LICENSE` for more information.

app/main.py

@@ -14,6 +14,7 @@
 client = WebClient(token=slack_token)
 verifier = SignatureVerifier(signing_secret)
 
+
 @app.route("/slack/commands", methods=["POST"])
 def slack_commands():
     data = request.form
@@ -25,6 +26,7 @@ def slack_commands():
         })
     return "Unknown command", 200
 
+
 @app.route("/slack/prompt", methods=["POST"])
 def slack_prompt():
     data = request.form
@@ -42,5 +44,6 @@ def slack_prompt():
         "text": f"<@{user_id}> {response}"
     })
 
+
 if __name__ == "__main__":
     app.run(debug=True,port=3000)

scripts/bot_script.py

@@ -20,6 +20,7 @@
 DATA_DIR = ROOT_DIR / "data"
 STORAGE_DIR = ROOT_DIR / "storage"
 
+
 async def build_and_save_index():
     """Builds the index from PDF files and saves it."""
     print("Loading PDF documents...")
@@ -39,6 +40,7 @@ async def build_and_save_index():
     index.storage_context.persist(persist_dir=STORAGE_DIR)
     return index
 
+
 async def load_or_build_index():
     """Loads existing index or builds a new one if not found."""
     if os.path.exists(STORAGE_DIR) or False:
@@ -49,6 +51,7 @@ async def load_or_build_index():
         index = await build_and_save_index()
     return index
 
+
 async def main():
     # Step 1: Setup global Settings
     Settings.embed_model = OpenAIEmbedding(api_key=os.getenv("OPENAI_API_KEY"))

services/slack_service.py

@@ -18,43 +18,45 @@
 client = WebClient(token=slack_token)
 verifier = SignatureVerifier(signing_secret)
 
+
 def get_rag_response(query: str) -> str:
     """
     Query the agent service and get response
     """
     import requests
     import json
-    
+
     # Configure the endpoint URL - adjust host/port as needed
     agent_service_url = f"{os.getenv("AGENT_SERVICE_URL")}/query"
-    
+
     try:
         # Make POST request to the agent service
         response = requests.post(
             agent_service_url,
             json={"query": query},
             headers={"Content-Type": "application/json"}
         )
-        
+
         # Check if request was successful
         response.raise_for_status()
-        
+
         # Parse JSON response
         result = response.json()
-        
+
         # Return the response text
         return result.get("response", "No response received from agent")
-        
+
     except requests.exceptions.RequestException as e:
         # Handle connection errors
         print(f"Error querying agent service: {e}")
         return f"Sorry, I couldn't reach the knowledge base. Error: {str(e)}"
-    
+
     except json.JSONDecodeError:
         # Handle invalid JSON response
         print(f"Invalid response from agent service: {response.text}")
         return "Sorry, I received an invalid response from the knowledge base."
 
+
 @app.route("/slack/commands", methods=["POST"])
 def slack_commands():
     data = request.form
@@ -66,6 +68,7 @@ def slack_commands():
         })
     return "Unknown command", 200
 
+
 @app.route("/slack/prompt", methods=["POST"])
 def slack_prompt():
     data = request.form
@@ -83,5 +86,6 @@ def slack_prompt():
         "text": f"<@{user_id}> {response}"
     })
 
+
 if __name__ == "__main__":
     app.run(debug=True,port=3000)