Daceflow AI
Unlock Possibilities →
📧 Connect with us: [email protected]

Link: https://youtu.be/1AmLD1aY7cM

This overview describes LangChain's agents in 9 minutes and is packed with examples and animations to get the main points across as simply as possible.

You will need Python and Pipenv

You will need

• OpenAI key (LLM)
• SERPApi key (Browsing the web)

Your .env file should look like

OPENAI_API_KEY="XXXXX"
SERPAPI_API_KEY="XXXXX"

You will need a credentials.json file to let the agent send emails. You can create yours at https://developers.google.com/gmail/api/quickstart/python#authorize_credentials_for_a_desktop_application

1. Clone the repository:

git clone https://github.com/edrickdch/langchain-agents

2. Navigate to the project directory:

cd langchain-agents

3. Install the required dependencies using Pipenv:

pipenv install

4. Activate the Pipenv shell:

pipenv shell

5. Create a .env file in the top-directory with the following:

OPENAI_API_KEY="XXXXX"
SERPAPI_API_KEY="XXXXX"

6. (OPTIONAL: For the sales report agent) Populate the DB

python src/agent/sales-report-db-starter.py

7. (OPTIONAL: For the sales report agent) Run the agent

python src/agent/sales-report-agent.py

• LangChain: https://python.langchain.com/docs/modules/agents/
• ReAct Paper: https://arxiv.org/abs/2210.03629
• MRKL Paper: https://www.ai21.com/blog/jurassic-x-crossing-the-neuro-symbolic-chasm-with-the-mrkl-system
• Self-ask with search Paper: https://python.langchain.com/docs/modules/agents/agent_types/self_ask_with_search
• BabyAGI: https://github.com/yoheinakajima/babyagi
• Generative Agents Paper: https://arxiv.org/abs/2304.03442
• Plan-and-solver Paper: https://github.com/AGI-Edgerunners/Plan-and-Solve-Prompting

👇 Subscribe to the newsletter if you're interested in building more AI applications

https://practical-ai-builder.beehiiv.com/