Using the Pythagora VS Code extension with your own API key

Note: This is only applicable to users who want to use Pythagora with their own API key. Pythagora Pro users do not need to follow the steps below, as no API key is required for Pro users.

To setup the Pythagora VS Code extension using your own API key:

1 - Via the command line, clone the GPT Pilot repo using the command: git clone https://github.com/Pythagora-io/gpt-pilot.git

2 - Inside the parent directory where GPT Pilot was cloned, go to GPT Pilot's directory: cd gpt-pilot

3 - Create a virtual environment: python3 -m venv venv

4 - Activate the virtual environment: source venv/bin/activate. For Windows users: venv\Scripts\activate

5 - Install the dependencies: pip install -r requirements.txt

6 - Copy the example-config.json file to create a config.json file: cp example-config.json config.json

7 - Edit the new config.json file to include the LLM you want to use and remove the other LLMs that you won't use.

8 - Add your own API keys, etc to the config.json file:

For OpenAI, the llm section of the config.json file should have:

"llm": {
  "openai": {
    "base_url": "https://api.openai.com/v1/",
    "api_key": "sk-your-api-key",

For Anthropic:

"llm": {
  "anthropic": {
    "base_url": "https://api.anthropic.com/",
    "api_key": "sk-your-api-key",

9 - Update the all of the agents in the agent section of the config.json file so the provider and model properties contain the correct LLM and model. An example of one agent using Anthropic (note: there may be more than one agent):

"agent": {
    "default": {
      "provider": "anthropic",
      "model": "claude-3-5-sonnet-20240620",
      "temperature": 0.5
    },

Note: Double-check the LLM's official documentation to ensure you have the correct model number. Click here for the OpenAI documentation. And here for Anthropic's docs. The wiki's Integrations Guides section has additional config examples for local LLM's, etc.

10 - Remove any commented out code from the config.json file.

11 - Inside of VS Code, go to the Pythagora extension's settings page by clicking on the gear icon. Make sure the GPT Pilot path is pointing to the directory containing the cloned repo.

For example, if your parent directory is called projects and you clone the GPT Pilot repo inside of projects, the projects directory will contain a subdirectory called gpt-pilot. In the Pythagora extension's settings, the GPT Pilot path section will need to be changed to the projects directory.

12 - Restart VS Code for the changes to be reflected on the Pythagora settings page.

13 - That's it! Pythagora's settings page should show that you're using your own API key (e.g. You're using your own Anthropic key.).