A curated collection of practical AI projects implementing OCR systems, RAG, AI agents, and other AI use cases.
added Agentic RAG with Firecrawl + Qwen
Joel James796bc357d3d0
diff --git a/agentic_rag/README.md b/agentic_rag_with_o-3-mini_and_duckduckgo/README.mdsimilarity index 89%rename from agentic_rag/README.mdrename to agentic_rag_with_o-3-mini_and_duckduckgo/README.mdindex c1bee0b..342451e 100644--- a/agentic_rag/README.md+++ b/agentic_rag_with_o-3-mini_and_duckduckgo/README.md@@ -1,4 +1,4 @@-# PDF RAG Agent with Milvus, Agno and OpenAI+# PDF RAG Agent using Milvus, Agno, OpenAI & DuckDuckGoA Streamlit application demonstrating a Retrieval-Augmented Generation (RAG) workflow using OpenAI (`gpt-4o-mini`), Agno, Milvus, and DuckDuckGo for web search.@@ -22,13 +22,16 @@ A Streamlit application demonstrating a Retrieval-Augmented Generation (RAG) wor1. **Clone Repository & Navigate:**```bashgit clone https://github.com/Sumanth077/awesome-ai-apps-and-agents.git- cd awesome-ai-apps-and-agents/agentic_rag++ cd awesome-ai-apps-and-agents/agentic_rag_with_o-3-mini_and_duckduckgo```2. **Environment & Dependencies:**```bashpython3 -m venv venv- source venv/bin/activate # Or .env\Scripts\activate on Windows++ source venv/bin/activate # or venv\Scripts\activate on Windows+pip install -r requirements.txt```diff --git a/agentic_rag/app.py b/agentic_rag_with_o-3-mini_and_duckduckgo/app.pysimilarity index 100%rename from agentic_rag/app.pyrename to agentic_rag_with_o-3-mini_and_duckduckgo/app.pydiff --git a/agentic_rag/requirements.txt b/agentic_rag_with_o-3-mini_and_duckduckgo/requirements.txtsimilarity index 100%rename from agentic_rag/requirements.txtrename to agentic_rag_with_o-3-mini_and_duckduckgo/requirements.txtdiff --git a/agentic_rag_with_qwen_and_firecrawl/README.md b/agentic_rag_with_qwen_and_firecrawl/README.mdnew file mode 100644index 0000000..6d1044f--- /dev/null+++ b/agentic_rag_with_qwen_and_firecrawl/README.md@@ -0,0 +1,72 @@+# PDF RAG Agent with Milvus, Agno, Qwen and FireCrawl++A Streamlit application demonstrating a Retrieval-Augmented Generation (RAG) workflow using Qwen3:8b, Agno, Milvus, and FireCrawl for crawling through relevant websites.++## Features++- Upload PDFs & process PDF documents.+- Generate BGE embeddings (`bge-small-en-v1.5`).+- Uses Qwen model (via Ollama) for LLM responses.+- Milvus vector store for semantic search.+- Firecrawl API integration for structured and up-to-date web search.+- Interactive Streamlit chat interface.++## Prerequisites++- Python 3.8++- FireCrawl API Key+- Docker (for Milvus)+- Running Milvus Instance++## Setup++1. **Clone Repository & Navigate:**+ ```bash+ git clone https://github.com/Sumanth077/awesome-ai-apps-and-agents.git++ cd awesome-ai-apps-and-agents/agentic_rag_with_qwen_and_firecrawl+ ```++2. **Environment & Dependencies:**+ ```bash+ python3 -m venv venv++ source venv/bin/activate # Or venv\Scripts\activate on Windows++ pip install -r requirements.txt+ ```++3. **FireCrawl API Key:**+ Set the `FIRECRAWL_API_KEY` environment variable or add it to your Streamlit secrets (`.streamlit/secrets.toml`).++4. **Start Milvus:**+ Follow the official Milvus guide to start a standalone instance using Docker:+ [https://milvus.io/docs/install_standalone-docker.md](https://milvus.io/docs/install_standalone-docker.md)+ *Ensure it's accessible (usually `http://localhost:19530`).*++## Usage++1. **Ensure Milvus is running.**+2. **Run the app:**+ ```bash+ streamlit run app.py+ ```+3. Open the provided URL (e.g., `http://localhost:8501`).+4. Upload a PDF, click "Process Document", and start chatting!++## How it Works++- PDF text is chunked and embedded using `bge-small-en-v1.5` from SentenceTransformers.+- Embeddings are stored in Milvus for semantic retrieval.+- User query triggers the Agno Agent:+ - Searches Milvus vector store for relevant context.+ - Falls back to Firecrawl for live web results if context is weak or missing.+ - Uses the locally hosted Qwen model via Ollama to synthesize a final response.++## Contributing++Contributions, issues, and feature requests are welcome. Please feel free to submit a Pull Request or open an issue.++## License++This project is licensed under the MIT License. See the LICENSE file for details (if one exists).\ No newline at end of filediff --git a/agentic_rag_with_qwen_and_firecrawl/app.py b/agentic_rag_with_qwen_and_firecrawl/app.pynew file mode 100644index 0000000..727374c--- /dev/null+++ b/agentic_rag_with_qwen_and_firecrawl/app.py@@ -0,0 +1,195 @@+import streamlit as st+from agno.models.ollama import Ollama+from agno.agent import Agent+from agno.knowledge.pdf import PDFKnowledgeBase+from agno.vectordb.milvus import Milvus+from agno.tools.firecrawl import FirecrawlTools+import os+from dotenv import load_dotenv+import base64+from sentence_transformers import SentenceTransformer++# Load embedder+embedding_model = SentenceTransformer("BAAI/bge-small-en-v1.5")++class CustomEmbedder:+ def __init__(self):+ self.model = SentenceTransformer("BAAI/bge-small-en-v1.5")+ self.dimensions = 384 # Required by agno Milvus wrapper++ def get_embedding(self, text: str) -> list[float]:+ return self.model.encode([text])[0].tolist()++ def embed(self, texts: list[str]) -> list[list[float]]:+ return self.model.encode(texts, show_progress_bar=False).tolist()++ def get_embedding_and_usage(self, texts: list[str]):+ embeddings = self.embed(texts)+ usage = {+ "input_tokens": sum(len(t.split()) for t in texts),+ "output_vectors": len(embeddings),+ }+ return embeddings, usage++# Set Streamlit page config+st.set_page_config(+ page_title="PDF RAG Agent (Local R1)",+ page_icon="π",+ layout="wide"+)++# Custom CSS styling+st.markdown("""+<style>+.stApp { color: #FAFAFA; }+.stSidebar > div:first-child {+ background-image: linear-gradient(to bottom, #262936, #1e202a);+}+[data-testid="stChatMessage"] {+ background-color: rgba(74, 74, 106, 0.4);+ border-radius: 0.5rem;+ padding: 1rem;+ margin-bottom: 0.5rem;+}+[data-testid="stChatMessage"] p { color: inherit; }+.pdf-preview-container {+ border: 1px solid #4A4A6A;+ border-radius: 0.5rem;+ padding: 0.5rem;+ background-color: #262936;+ margin-bottom: 1rem;+}+</style>+""", unsafe_allow_html=True)++load_dotenv()++firecrawl_api_key = os.getenv("FIRECRAWL_API_KEY")++# Milvus Vector DB+MILVUS_URI = os.getenv("MILVUS_URI", "http://localhost:19530")+COLLECTION_NAME = "rag_documents_local"++# Configure Milvus with the custom embedder+vector_db = Milvus(+ collection=COLLECTION_NAME,+ uri=MILVUS_URI,+ embedder=CustomEmbedder()+)++# Create Agent+def get_rag_agent(knowledge_base: PDFKnowledgeBase, model_id="qwen3:8b", debug_mode=True):+ model = Ollama(id=model_id)+ instructions = [+ "1. Knowledge Base Search:",+ " - ALWAYS start by searching the knowledge base using search_knowledge_base tool",+ " - Analyze ALL returned documents thoroughly before responding",+ " - If multiple documents are returned, synthesize the information coherently",+ "2. External Search:",+ " - If knowledge base search yields insufficient results, use duckduckgo_search",+ " - Focus on reputable sources and recent information",+ " - Cross-reference information from multiple sources when possible",+ "3. Citation Precision:",+ " - Reference page numbers and section headers",+ " - Distinguish between main content and appendices",+ "4. Response Quality:",+ " - Provide specific citations and sources for claims",+ " - Structure responses with clear sections and bullet points when appropriate",+ " - Include relevant quotes from source materials",+ " - Avoid hedging phrases like 'based on my knowledge' or 'depending on the information'",+ "5. Response Structure:",+ " - Use markdown for formatting technical content",+ " - Create bullet points for lists found in documents",+ " - Preserve important formatting from original PDF",+ "6. User Interaction:",+ " - Ask for clarification if the query is ambiguous",+ " - Break down complex questions into manageable parts",+ " - Proactively suggest related topics or follow-up questions",+ "7. Error Handling:",+ " - If no relevant information is found, clearly state this",+ " - Suggest alternative approaches or questions",+ " - Be transparent about limitations in available information",+ ]++ return Agent(+ model=model,+ knowledge=knowledge_base,+ description="You are a helpful Agent called 'Agentic RAG' assisting with questions about a PDF document.", # Updated description+ instructions=instructions,+ search_knowledge=True,+ markdown=True,+ tools=[FirecrawlTools(api_key=firecrawl_api_key, scrape=False, crawl=True)],+ show_tool_calls=True,+ add_datetime_to_instructions=False,+ debug_mode=debug_mode,+ )++# PDF Viewer+def display_pdf_preview(pdf_path):+ try:+ with open(pdf_path, "rb") as f:+ base64_pdf = base64.b64encode(f.read()).decode("utf-8")+ st.markdown(f'<iframe src="data:application/pdf;base64,{base64_pdf}" width="100%" height="300px"></iframe>', unsafe_allow_html=True)+ except Exception as e:+ st.error(f"Error displaying PDF: {e}")++# Session State Init+for key in ["messages", "document_loaded", "agent", "knowledge_base"]:+ if key not in st.session_state:+ st.session_state[key] = [] if key == "messages" else None++# Sidebar: Upload PDF+with st.sidebar:+ st.markdown("## Upload PDF")+ uploaded_file = st.file_uploader("Choose a PDF", type=["pdf"])++ if uploaded_file:+ pdf_path = "temp_uploaded.pdf"+ with open(pdf_path, "wb") as f:+ f.write(uploaded_file.read())++ display_pdf_preview(pdf_path)++ if not st.session_state.document_loaded:+ if st.button("Process Document"):+ with st.spinner("Indexing document..."):+ kb = PDFKnowledgeBase(path=pdf_path, vector_db=vector_db)++ kb.load(recreate=True)+ agent = get_rag_agent(kb)++ st.session_state.knowledge_base = kb+ st.session_state.agent = agent+ st.session_state.document_loaded = True+ st.session_state.messages = []+ st.rerun()++ else:+ st.markdown("β Document indexed!")++# Main Chat Area+st.markdown("# Chat with your PDF")+if not st.session_state.document_loaded:+ st.info("π Upload and process a PDF to begin.")+else:+ for msg in st.session_state.messages:+ with st.chat_message(msg["role"]):+ st.markdown(msg["content"])++ if prompt := st.chat_input("Ask something about your document..."):+ st.session_state.messages.append({"role": "user", "content": prompt})+ with st.chat_message("user"):+ st.markdown(prompt)++ with st.chat_message("assistant"):+ msg_box = st.empty()+ try:+ response = st.session_state.agent.run(prompt)+ content = response.content+ msg_box.markdown(content)+ except Exception as e:+ content = f"Error: {e}"+ st.error(content)++ st.session_state.messages.append({"role": "assistant", "content": content})+ st.rerun()\ No newline at end of filediff --git a/agentic_rag_with_qwen_and_firecrawl/requirements.txt b/agentic_rag_with_qwen_and_firecrawl/requirements.txtnew file mode 100644index 0000000..71b00ba--- /dev/null+++ b/agentic_rag_with_qwen_and_firecrawl/requirements.txt@@ -0,0 +1,94 @@+agno==1.5.10+aiohappyeyeballs==2.6.1+aiohttp==3.12.12+aiosignal==1.3.2+altair==5.5.0+annotated-types==0.7.0+anyio==4.9.0+async-timeout==5.0.1+attrs==25.3.0+blinker==1.9.0+cachetools==5.5.2+certifi==2025.4.26+charset-normalizer==3.4.2+click==8.2.1+colorama==0.4.6+docstring_parser==0.16+exceptiongroup==1.3.0+filelock==3.18.0+firecrawl-py==2.8.0+frozenlist==1.7.0+fsspec==2025.5.1+gitdb==4.0.12+GitPython==3.1.44+grpcio==1.67.1+h11==0.16.0+httpcore==1.0.9+httpx==0.28.1+huggingface-hub==0.32.4+idna==3.10+Jinja2==3.1.6+joblib==1.5.1+jsonschema==4.24.0+jsonschema-specifications==2025.4.1+markdown-it-py==3.0.0+MarkupSafe==3.0.2+mdurl==0.1.2+mpmath==1.3.0+multidict==6.4.4+narwhals==1.42.0+nest-asyncio==1.6.0+networkx==3.4.2+numpy==2.2.6+ollama==0.5.1+packaging==24.2+pandas==2.3.0+pillow==11.2.1+propcache==0.3.2+protobuf==6.31.1+pyarrow==20.0.0+pydantic==2.11.5+pydantic-settings==2.9.1+pydantic_core==2.33.2+pydeck==0.9.1+Pygments==2.19.1+pymilvus==2.5.11+pypdf==5.6.0+python-dateutil==2.9.0.post0+python-dotenv==1.1.0+python-multipart==0.0.20+pytz==2025.2+PyYAML==6.0.2+referencing==0.36.2+regex==2024.11.6+requests==2.32.4+rich==14.0.0+rpds-py==0.25.1+safetensors==0.5.3+scikit-learn==1.7.0+scipy==1.15.3+sentence-transformers==4.1.0+shellingham==1.5.4+six==1.17.0+smmap==5.0.2+sniffio==1.3.1+streamlit==1.45.1+sympy==1.14.0+tenacity==9.1.2+threadpoolctl==3.6.0+tokenizers==0.21.1+toml==0.10.2+tomli==2.2.1+torch==2.7.1+tornado==6.5.1+tqdm==4.67.1+transformers==4.52.4+typer==0.16.0+typing-inspection==0.4.1+typing_extensions==4.14.0+tzdata==2025.2+ujson==5.10.0+urllib3==2.4.0+watchdog==6.0.0+websockets==15.0.1+yarl==1.20.1