ToolCreate
Payload used to create Tool. Required fields are name, description, type.
alias object
- string
- null
Description of the tool's functionality
Tool type
Possible values: [python, agent, workflow]
collection_id object
Identifier of the containing collection.
- string
- null
Possible values: non-empty
folder_id object
- string
- null
tags object
- string[]
- null
function object
Python function schema
- PythonFunctionSchema
- null
inputs object
{}outputs object
{}[]decorators object[]
Decorator bindings declared in configuration. Configured bindings take precedence over source bindings with the same kind and name; an empty list uses source bindings only.
Decorator name, such as use_connector or use_llm. The name must identify a supported decorator.
Binding variable name injected into the namespace
arguments object
Decorator arguments, e.g. {'ref':...} or {'model':...}
Decorator arguments, e.g. {'ref':...} or {'model':...}
ref object
Referenced agent or pipeline name (for agent/workflow types)
- string
- null
Whether this is a built-in tool
falseWhether this tool needs a browser sidecar at agent deploy
falseexamples object[]
Worked call examples surfaced to the LLM as few-shot guidance
description object
- string
- null
arguments object필수
options object
- object
- null
{
"name": "string",
"alias": "string",
"description": "string",
"type": "python",
"collection_id": "string",
"folder_id": "string",
"tags": [
"string"
],
"function": {
"source": "string",
"inputs": {},
"outputs": {},
"dependencies": [
"string"
],
"decorators": [
{
"decorator": "string",
"name": "string",
"arguments": {}
}
]
},
"ref": "string",
"is_builtin": false,
"require_browser": false,
"examples": [
{
"description": "string",
"arguments": {}
}
],
"options": {}
}