Dify API Key

Consider this scenario:

You’re using OpenAI and obtain your API Key from https://platform.openai.com/api-keys.

You set up two different SYSTEM prompts in the PLAYGROUND:

1. One that instructs GPT to act like a cat
2. Another that instructs GPT to act like a dog

By clicking the , you receive two Python code snippets - one for the cat behavior and one for the dog behavior.

A sample code might look like this:

from openai import OpenAI
client = OpenAI(api_key="your_api_key_here")

response = client.chat.completions.create(
  model="gpt-4o",
  messages=[
    {
      "role": "system",
      "content": [
        {
          "type": "text",
          "text": "I need you to reply like a cat"
        }
      ]
    }
  ],
  temperature=1,
  max_tokens=256,
  top_p=1,
  frequency_penalty=0,
  presence_penalty=0,
  response_format={
    "type": "text"
  }
)

You create two local Python files: cat_simulator.py and dog_simulator.py.

You can run these files in the terminal using python cat_simulator.py and python dog_simulator.py, or execute them in your preferred environment.

Key Components

This project has three crucial elements:

1. “your_api_key_here” - The API Key that grants access to the models.
2. “cat_simulator” - Represents a Dify APP “A”.
3. “dog_simulator” - Represents a Dify APP “B”.

A Dify APP is constructed using both your LLM provider’s model and your custom workflow, creating a combination like: cat_simulator&your_api_key_here.

After clicking Create new Secret Key, you receive an “API Secret key”, let’s call it Dify-API-Key-A.

Similarly, you can have dog_simulator&your_api_key_here, and Dify-API-Key-B.

When using the API with “Dify-API-Key-A”, will it interact with the “dog_simulator”?