General Specs

This article will briefly introduce common structures in plugin development. During development, it is strongly recommended to read this alongside Basic Concepts of Plugin Development and the Developer Cheatsheet for a better understanding of the overall architecture.

Path Specifications

When filling in file paths in Manifest or any yaml files, follow these two specifications depending on the type of file:

• If the target file is a multimedia file such as an image or video, for example when filling in the plugin’s icon, you should place these files in the _assets folder under the plugin’s root directory.
• If the target file is a regular text file, such as .py or .yaml code files, you should fill in the absolute path of the file within the plugin project.

Common Structures

When defining plugins, there are some data structures that can be shared between tools, models, and Endpoints. These shared structures are defined here.

I18nObject

I18nObject is an internationalization structure that conforms to the IETF BCP 47 standard. Currently, four languages are supported:

en_US · string

English (United States)

zh_Hans · string

Simplified Chinese

ja_JP · string

Japanese

pt_BR · string

Portuguese (Brazil)

ProviderConfig

ProviderConfig is a common provider form structure, applicable to both Tool and Endpoint

name · string

Form item name

label · I18nObject · required

Display labels following IETF BCP 47 standard

type · provider_config_type · required

Form field type - determines how the field will be rendered in the UI

scope · provider_config_scope

Optional range specification, varies based on the value of type

required · boolean

Whether the field cannot be empty

default · any

Default value, only supports basic types: float, int, string

options · array[provider_config_option]

Available options, only used when type is select

helper · object

Help document link label, following IETF BCP 47

url · string

Help document link

placeholder · object

Placeholder text in multiple languages, following IETF BCP 47

ProviderConfigOption(object)

value · string · required

The value of the option

label · object · required

Display label for the option, following IETF BCP 47

ProviderConfigType(string)

secret-input · string

Configuration information that will be encrypted

text-input · string

Plain text input field

select · string

Dropdown selection field

boolean · boolean

Switch/toggle control

model-selector · object

Model configuration selector, including provider name, model name, model parameters, etc.

app-selector · object

Application ID selector

tool-selector · object

Tool configuration selector, including tool provider, name, parameters, etc.

dataset-selector · string

Dataset selector (TBD)

ProviderConfigScope(string)

When type is model-selector:

all · string

All model types

llm · string

Large Language Models only

text-embedding · string

Text embedding models only

rerank · string

Reranking models only

tts · string

Text-to-speech models only

speech2text · string

Speech-to-text models only

moderation · string

Content moderation models only

vision · string

Vision models only

When type is app-selector:

all · string

All application types

chat · string

Chat applications only

workflow · string

Workflow applications only

completion · string

Completion applications only

When type is tool-selector:

all · string

All tool types

plugin · string

Plugin tools only

api · string

API tools only

workflow · string

Workflow tools only

ModelConfig

provider · string

Model provider name containing plugin_id, in the form of langgenius/openai/openai

model · string

Specific model name

model_type · enum

Enumeration of model types, refer to the Model Design Rules document

NodeResponse

inputs · dict

Variables that are finally input to the node

outputs · dict

Output results of the node

process_data · dict

Data generated during node execution

ToolSelector

provider_id · string

Tool provider name

tool_name · string

Tool name

tool_description · string

Tool description

tool_configuration · dict[string, any]

Tool configuration information

tool_parameters · dict[string, dict]

Parameters that need LLM reasoning

name · string

Parameter name

type · string

Parameter type

required · boolean

Whether the parameter is required

description · string

Parameter description

default · any

Default value

options · array[string]

Available options for the parameter

Related Resources

• Basic Concepts of Plugin Development - Comprehensive understanding of Dify plugin development
• Developer Cheatsheet - Quick reference for common commands and concepts in plugin development
• Tool Plugin Development Details - Understanding how to define plugin information and the tool plugin development process
• Model Design Rules - Understanding the standards for model configuration

---

Edit this page | Report an issue