Could the Dify knowledge base be implemented this way?

[lemonit-eric-mao]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

Title: Can we implement Dify’s knowledge base this way?

I think the current approach of Dify’s knowledge base is not flexible enough. For example, I might want to change the way documents are processed, or choose a niche database based on my specific business needs.

In some cases, I may need to dynamically create vector databases depending on the business requirements, or perform hybrid searches with multiple vector columns based on different retrieval scenarios.

Traditionally, knowledge base products are customized by developers through Python code, often using a component-based approach during development.

Given these scenarios, I have a new idea:
what if the knowledge base itself is a workflow?
If so, could we use the existing workflow engine to customize and build a knowledge base?

Here’s my proposed approach:

Create Multiple Workflows for Future Use

Create a Database

Insert Data

Vector Search

Query

Related DSL Files

DSL.zip

2. Additional context or comments

No response

3. Can you help us with this feature?

• I am interested in contributing to this feature.

[lemonit-eric-mao]

My idea is that the Dify knowledge base should invoke such workflows, rather than encapsulating all the functionality inside a fixed implementation.
Over-encapsulation makes it difficult to adapt to the diverse and complex business scenarios found across different enterprises.