Developers can extend module capabilities through the API extension module. Currently supported module extensions include:
moderation
external_data_tool
Before extending module capabilities, prepare an API and an API Key for authentication, which can also be automatically generated by Dify. In addition to developing the corresponding module capabilities, follow the specifications below so that Dify can invoke the API correctly.
Add API Extension
API Specifications
Dify will invoke your API according to the following specifications:
Header
Header
Value
Desc
Content-Type
application/json
The request content is in JSON format.
Authorization
Bearer {api_key}
The API Key is transmitted as a token. You need to parse the api_key and verify if it matches the provided API Key to ensure API security.
Request Body
API Response
Check
When configuring API-based Extension in Dify, Dify will send a request to the API Endpoint to verify the availability of the API. When the API Endpoint receives point=ping, the API should return result=pong, as follows:
Header
Request Body
Expected API response
\
For Example
Here we take the external data tool as an example, where the scenario is to retrieve external weather information based on the region as context.
API Specifications
POST https://fake-domain.com/api/dify/receive
Header
Request Body
API Response
Code demo
The code is based on the Python FastAPI framework.
Install dependencies.
Write code according to the interface specifications.
Launch the API service.
The default port is 8000. The complete address of the API is: http://127.0.0.1:8000/api/dify/receivewith the configured API Key '123456'.
Configure this API in Dify.
Select this API extension in the App.
When debugging the App, Dify will request the configured API and send the following content (example):
API Response：
Local debugging
Since Dify's cloud version can't access internal network API services, you can use Ngrok to expose your local API service endpoint to the public internet for cloud-based debugging of local code. The steps are:
Visit the Ngrok official website at https://ngrok.com, register, and download the Ngrok file.
After downloading, go to the download directory. Unzip the package and run the initialization script as instructed:
Check the port of your local API service.
Run the following command to start:
Upon successful startup, you'll see something like the following:
Find the 'Forwarding' address, like the sample domain https://177e-159-223-41-52.ngrok-free.app, and use it as your public domain.
For example, to expose your locally running service, replace the example URL http://127.0.0.1:8000/api/dify/receive with https://177e-159-223-41-52.ngrok-free.app/api/dify/receive.
Now, this API endpoint is accessible publicly. You can configure this endpoint in Dify for local debugging. For the configuration steps, consult the appropriate documentation or guide.
Deploy API extension with Cloudflare Workers
We recommend that you use Cloudflare Workers to deploy your API extension, because Cloudflare Workers can easily provide a public address and can be used for free.
{
"point": string, // Extension point, different modules may contain multiple extension points
"params": {
... // Parameters passed to each module's extension point
}
}
{
... // For the content returned by the API, see the specific module's design specifications for different extension points.
}
from fastapi import FastAPI, Body, HTTPException, Header
from pydantic import BaseModel
app = FastAPI()
class InputData(BaseModel):
point: str
params: dict
@app.post("/api/dify/receive")
async def dify_receive(data: InputData = Body(...), authorization: str = Header(None)):
"""
Receive API query data from Dify.
"""
expected_api_key = "123456" # TODO Your API key of this API
auth_scheme, _, api_key = authorization.partition(' ')
if auth_scheme.lower() != "bearer" or api_key != expected_api_key:
raise HTTPException(status_code=401, detail="Unauthorized")
point = data.point
# for debug
print(f"point: {point}")
if point == "ping":
return {
"result": "pong"
}
if point == "app.external_data_tool.query":
return handle_app_external_data_tool_query(params=data.params)
# elif point == "{point name}":
# TODO other point implementation here
raise HTTPException(status_code=400, detail="Not implemented")
def handle_app_external_data_tool_query(params: dict):
app_id = params.get("app_id")
tool_variable = params.get("tool_variable")
inputs = params.get("inputs")
query = params.get("query")
# for debug
print(f"app_id: {app_id}")
print(f"tool_variable: {tool_variable}")
print(f"inputs: {inputs}")
print(f"query: {query}")
# TODO your external data tool query implementation here,
# return must be a dict with key "result", and the value is the query result
if inputs.get("location") == "London":
return {
"result": "City: London\nTemperature: 10°C\nRealFeel®: 8°C\nAir Quality: Poor\nWind Direction: ENE\nWind "
"Speed: 8 km/h\nWind Gusts: 14 km/h\nPrecipitation: Light rain"
}
else:
return {"result": "Unknown city"}
