On this page
Dify: agent building features and setup
A platform for building AI applications, agents, and workflows that draw on organizational knowledge. It combines a workflow editor, knowledge source preparation, and several publishing options.
Tool by the Otomatzia editorial team · source content checked on
Overview
Dify helps turn model instructions into a workflow that can be inspected and tested: receive data, retrieve information, use a tool, and return an answer. Its knowledge system includes preparing and indexing documents and sources, not just a field for uploading a file. You can publish an application, embed it on a website, or access it through an API. Hosting options vary, so suitability for an organization also depends on licensing, maintenance, and where user data is processed.
Features
- Workflow Studio for combining logic, models, tools, and human approval.
- Knowledge workflows for extracting, cleaning, splitting, and indexing sources.
- Publish as an app, API, embed, or MCP-compatible tool.
Business uses
- An internal assistant that finds answers in procedure documents and presents relevant information.
- A workflow that processes a document, extracts data, and requests approval before transferring it to a system.
Getting started
- Choose cloud or self-hosting and select a suitable model provider.
- Upload a small document set and check whether retrieval finds the right passage.
- Publish to a test environment and examine missing answers, response times, and run history.
What to check
- For the self-hosted edition, check the license terms, especially before building a multi-tenant product or changing the branding.
- Hosting Dify locally does not guarantee that all data stays local if a cloud model is connected.
Sources to check next
Vendor features and prices change. Confirm details on the linked official pages before you rely on them.