Assistants Cookbook

Phidata Assistants add memory, knowledge and tools to LLMs. Let's test out a few examples.

• Create a virtual environment

python3 -m venv ~/.venvs/aienv
source ~/.venvs/aienv/bin/activate

• Install libraries

pip install -U phidata openai

Assistants

• Basic Assistant

python cookbook/assistants/basic.py

Assistants with Tools

• Data Analyst Assistant

pip install -U duckdb

python cookbook/assistants/data_analyst.py

• Web Search Assistant

pip install -U duckduckgo-search

python cookbook/assistants/web_search.py

• Python Assistant

python cookbook/assistants/python_assistant.py

Run PgVector

Install docker desktop first.

• Run using a helper script

./cookbook/run_pgvector.sh

• OR run using the docker run command

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

Assistants with Knowledge

• Install libraries

pip install -U sqlalchemy pgvector "psycopg[binary]" pypdf

• RAG Assistant

python cookbook/assistants/rag_assistant.py

• Autonomous Assistant

python cookbook/assistants/auto_assistant.py

Assistants with Storage, Knowledge & Tools

• PDF Assistant

python cookbook/assistants/pdf_assistant.py