Merge branch 'main' of https://github.com/phidatahq/phidata

.github/workflows/stale-issues.yml

@@ -0,0 +1,31 @@
+name: Close Stale Issues
+
+on:
+  schedule:
+    - cron: "0 0 * * *"
+  workflow_dispatch:
+
+jobs:
+  close_stale_issues:
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+
+    steps:
+      - name: Close stale issues
+        uses: actions/stale@v5
+        with:
+          stale-issue-message: >
+            This issue has been marked as stale because it has been inactive
+            for more than 14 days. Please update this issue or it will be
+            automatically closed.
+          days-before-issue-stale: 14
+          days-before-issue-close: 0
+          stale-issue-label: stale
+
+          # Disable PR processing
+          days-before-pr-stale: -1
+          days-before-pr-close: -1
+
+          # Only process issues, not PRs
+          only-issue-labels: "*"

cookbook/agents_101/agent_ui.py → cookbook/agents_101/04_agent_ui.py

@@ -38,4 +38,4 @@
 app = Playground(agents=[finance_agent, web_agent, agent_team]).get_app()
 
 if __name__ == "__main__":
-    serve_playground_app("agent_ui:app", reload=True)
+    serve_playground_app("04_agent_ui:app", reload=True)

cookbook/agents_101/README.md

@@ -21,22 +21,22 @@ export OPENAI_API_KEY=***
 pip install -U openai duckduckgo-search duckdb yfinance lancedb tantivy pypdf sqlalchemy 'fastapi[standard]' phidata
 ```
 
-### 4. Web Search Agent
+### 4. Run the Web Search Agent
 
 ```shell
-python cookbook/agents_101/web_search.py
+python cookbook/agents_101/01_web_search.py
 ```
 
-### 5. Finance Agent
+### 5. Run the Finance Agent
 
 ```shell
-python cookbook/agents_101/finance_agent.py
+python cookbook/agents_101/02_finance_agent.py
 ```
 
-### 6. RAG Agent
+### 6. Run the RAG Agent
 
 ```shell
-python cookbook/agents_101/rag_agent.py
+python cookbook/agents_101/03_rag_agent.py
 ```
 
 ### 7. Test in Agent UI
@@ -50,5 +50,5 @@ phi auth
 Run the Agent UI
 
 ```shell
-python cookbook/agents_101/agent_ui.py
+python cookbook/agents_101/04_agent_ui.py
 ```

cookbook/knowledge/combined_kb.py

@@ -0,0 +1,72 @@
+from pathlib import Path
+
+from phi.agent import Agent
+from phi.knowledge.csv import CSVKnowledgeBase
+from phi.knowledge.pdf import PDFKnowledgeBase, PDFUrlKnowledgeBase
+from phi.knowledge.website import WebsiteKnowledgeBase
+from phi.knowledge.combined import CombinedKnowledgeBase
+from phi.vectordb.pgvector import PgVector
+
+db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
+
+# Create CSV knowledge base
+csv_kb = CSVKnowledgeBase(
+    path=Path("data/csvs"),
+    vector_db=PgVector(
+        table_name="csv_documents",
+        db_url=db_url,
+    ),
+)
+
+# Create PDF URL knowledge base
+pdf_url_kb = PDFUrlKnowledgeBase(
+    urls=["https://phi-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
+    vector_db=PgVector(
+        table_name="pdf_documents",
+        db_url=db_url,
+    ),
+)
+
+# Create Website knowledge base
+website_kb = WebsiteKnowledgeBase(
+    urls=["https://docs.phidata.com/introduction"],
+    max_links=10,
+    vector_db=PgVector(
+        table_name="website_documents",
+        db_url=db_url,
+    ),
+)
+
+# Create Local PDF knowledge base
+local_pdf_kb = PDFKnowledgeBase(
+    path="data/pdfs",
+    vector_db=PgVector(
+        table_name="pdf_documents",
+        db_url=db_url,
+    ),
+)
+
+# Combine knowledge bases
+knowledge_base = CombinedKnowledgeBase(
+    sources=[
+        csv_kb,
+        pdf_url_kb,
+        website_kb,
+        local_pdf_kb,
+    ],
+    vector_db=PgVector(
+        table_name="combined_documents",
+        db_url=db_url,
+    ),
+)
+
+# Initialize the Agent with the combined knowledge base
+agent = Agent(
+    knowledge_base=knowledge_base,
+    add_references_to_prompt=True,
+)
+
+knowledge_base.load(recreate=False)
+
+# Use the agent
+agent.print_response("Ask me about something from the knowledge base", markdown=True)

cookbook/knowledge/csv_kb.py

@@ -0,0 +1,28 @@
+from pathlib import Path
+
+from phi.agent import Agent
+from phi.knowledge.csv import CSVKnowledgeBase
+from phi.vectordb.pgvector import PgVector
+
+db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
+
+
+knowledge_base = CSVKnowledgeBase(
+    path=Path("data/csvs"),
+    vector_db=PgVector(
+        table_name="csv_documents",
+        db_url=db_url,
+    ),
+    num_documents=5,  # Number of documents to return on search
+)
+# Load the knowledge base
+knowledge_base.load(recreate=False)
+
+# Initialize the Agent with the knowledge_base
+agent = Agent(
+    knowledge_base=knowledge_base,
+    add_references_to_prompt=True,
+)
+
+# Use the agent
+agent.print_response("Ask me about something from the knowledge base", markdown=True)

cookbook/knowledge/docx_kb.py

@@ -0,0 +1,27 @@
+from pathlib import Path
+
+from phi.agent import Agent
+from phi.vectordb.pgvector import PgVector
+from phi.knowledge.docx import DocxKnowledgeBase
+
+db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"
+
+# Create a knowledge base with the DOCX files from the data/docs directory
+knowledge_base = DocxKnowledgeBase(
+    path=Path("data/docs"),
+    vector_db=PgVector(
+        table_name="docx_documents",
+        db_url=db_url,
+    ),
+)
+# Load the knowledge base
+knowledge_base.load(recreate=False)
+
+# Create an agent with the knowledge base
+agent = Agent(
+    knowledge_base=knowledge_base,
+    add_references_to_prompt=True,
+)
+
+# Ask the agent about the knowledge base
+agent.print_response("Ask me about something from the knowledge base", markdown=True)

cookbook/knowledge/json_kb.py

@@ -8,7 +8,7 @@
 
 # Initialize the JSONKnowledgeBase
 knowledge_base = JSONKnowledgeBase(
-    path=Path("data/docs"),  # Table name: ai.json_documents
+    path=Path("data/json"),  # Table name: ai.json_documents
     vector_db=PgVector(
         table_name="json_documents",
         db_url=db_url,
@@ -18,7 +18,7 @@
 # Load the knowledge base
 knowledge_base.load(recreate=False)
 
-# Initialize the Assistant with the knowledge_base
+# Initialize the Agent with the knowledge_base
 agent = Agent(
     knowledge_base=knowledge_base,
     add_references_to_prompt=True,

cookbook/rag/README.md

@@ -0,0 +1,72 @@
+# Agentic RAG
+
+**RAG:** is a technique that allows an Agent to search for information to improve its responses. This directory contains a series of cookbooks that demonstrate how to build a RAG for the Agent.
+
+> Note: Fork and clone this repository if needed
+
+### 1. Create a virtual environment
+
+```shell
+python3 -m venv ~/.venvs/aienv
+source ~/.venvs/aienv/bin/activate
+```
+
+### 2. Install libraries
+
+```shell
+pip install -U openai sqlalchemy "psycopg[binary]" pgvector lancedb tantivy pypdf sqlalchemy "fastapi[standard]" phidata
+```
+
+### 3. Run PgVector
+
+> Install [docker desktop](https://docs.docker.com/desktop/install/mac-install/) first.
+
+- Run using a helper script
+
+```shell
+./cookbook/run_pgvector.sh
+```
+
+- OR run using the docker run command
+
+```shell
+docker run -d \
+  -e POSTGRES_DB=ai \
+  -e POSTGRES_USER=ai \
+  -e POSTGRES_PASSWORD=ai \
+  -e PGDATA=/var/lib/postgresql/data/pgdata \
+  -v pgvolume:/var/lib/postgresql/data \
+  -p 5532:5432 \
+  --name pgvector \
+  phidata/pgvector:16
+```
+
+### 4. Run the Traditional RAG with PgVector
+
+```shell
+python cookbook/rag/01_traditional_rag_pgvector.py
+```
+
+### 5. Run the Agentic RAG with PgVector
+
+```shell
+python cookbook/rag/02_agentic_rag_pgvector.py
+```
+
+### 6. Run the Traditional RAG with LanceDB
+
+```shell
+python cookbook/rag/03_traditional_rag_lancedb.py
+```
+
+### 7. Run the Agentic RAG with LanceDB
+
+```shell
+python cookbook/rag/04_agentic_rag_lancedb.py
+```
+
+### 8. Run the Agentic RAG on Agent UI
+
+```shell
+python cookbook/rag/05_agentic_rag_agent_ui.py
+```