Web Researcher

This Recipe shows an example of a web researcher chatbot.

Core Concepts

How to Configure Built-in Components

Agents

Conversational Agent

This uses the SimpleAgent template, but needs some customization to enable file uploads and support multiple LLMs.

Speech Agent

This uses the builtin AutonomousSpeechAgent to enable speech-to-text capabilities.

Web Researcher

This is the researcher agent. Customize the prompts, search parameters, etc… to suit your needs.

You will notice that enabled file upload on our AgentProcessingUnit’s primary action.

  actions:
    - name: "converse"
      description: "A copilot that engages with the user."
      allow_file_upload: true

We also have a list of available APUs in resources/apus.yaml.

  apus:
    - apu: MistralSmall
      title: Mistral Small
    - apu: MistralMedium
      title: Mistral Medium
    - apu: MistralLarge
...

We did not need to make any customization to support multimedia within the APU, this is turned on by default 🚀.

Try it out!

First let’s clone the Eidolon’s web researcher repository to your local machine and start your server.

git clone https://github.com/eidolon-ai/web-researcher.git
cd web-researcher

Then run the server and webui in dev mode, use the following command:

make docker-serve

⚠️ The first time you run this command, you may be prompted to enter credentials that the machine needs to run (ie, OpenAI API Key, Google CSE key, and Google CSE Token). These resources will be saved in the `.env` file in the project root.

If the server starts successfully, you should see the following output:

Starting Server...
INFO:     Started server process [34623]
INFO:     Waiting for application startup.
INFO - Building machine 'local_dev'
...
INFO - Server Started in 1.50s

Now Head over to the chatbot ui in your favorite browser and start chatting with your new agent.