john-adeojo/custom_agent_tutorial

Setting Up and Running Custom Agent Script

Prerequisites

1. Install Anaconda:
   Download Anaconda from https://www.anaconda.com/.

2. Create a Virtual Environment:

   conda create -n agent_env python=3.10 pip

3. Activate the Virtual Environment:

   conda activate agent_env

Clone and Navigate to the Repository

1. Clone the Repo:

   git clone https://github.com/john-adeojo/custom_agent_tutorial.git

2. Navigate to the Repo:

   cd /path/to/your-repo/custom_agent_tutorial

3. Install Requirements:

   pip install -r requirements.txt

Configure API Keys

1. Open the config.yaml: Update your keys, see next step for links to get your own API keys.

nano config.yaml

3. Enter API Keys:
   • Serper API Key: Get it from https://serper.dev/
   • OpenAI API Key: Get it from https://openai.com/

Run Your Query

python agent.py run

Then run your query

custom_agent_tutorial's Issues

Error: 'choices' key not found in response. Using empty content.

python agent.py run
Enter your query: https://inverbathrooms.ie/ extract : Contact informations headquarter or any location offices all phones with details any team member , managmenet member listed give me names and contact also all adressess including HQ, other locations , distributors make the result pretty read able and explain

gives error:

Error: 'choices' key not found in response. Using empty content.
Integration Agent:
Traceback (most recent call last):
File "/home/bc/Projects/custom_agent_tutorial/agent.py", line 166, in
agent.execute()
File "/home/bc/Projects/custom_agent_tutorial/agent.py", line 151, in execute
plan = self.run_planning_agent(query, plan=plan, outputs=outputs, feedback=response)
File "/home/bc/Projects/custom_agent_tutorial/agent.py", line 54, in run_planning_agent
content = response_dict['choices'][0]['message']['content']
KeyError: 'choices'

Using Ollama insted of ChatGPT

I came across your video 'Forget CrewAI & AutoGen, Build CUSTOM AI Agents!' and thought you made many good points, so I wanted to try the code myself. I don't have an OpenGPT account, so I wanted to try to adapt the code to use Ollama and llama3 instead. It all seemed quite manageable, but the problem I'm a bit stuck on / wondering how to get around or reimplement is the 'tool_calls' object that OpenGPT apparently has in its return JSON.

Do you have any suggestions or insights on how to best work around this? I tried following the 'solution' here (ollama/ollama-python#39 (comment)), but I can't get it to include 'tool_calls'. Any ideas or suggestions?