opencode/yao
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Refactor provider schemas to utilize bindata for enhanced localization support

Browse source 
- Updated Schema methods for various providers (OpenAI, Fastembed, MCP, OCR, Office, UTF8, Video, Vision, Whisper) to retrieve schemas from bindata, improving localization capabilities.
- Removed outdated JSON files for embedding, extractor, and fetcher providers, streamlining the data structure.
- Enhanced the overall provider management by ensuring schemas are dynamically loaded based on locale, facilitating better multi-language support.
This commit is contained in:
Max 2025-08-10 15:51:34 +08:00
parent 6ffc847382
commit eb1bf43d61
43 changed files with 4970 additions and 208 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

770
data/bindata.go
View file

File diff suppressed because one or more lines are too long

3
kb/providers/converters/mcp.go
View file

@ -5,6 +5,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -128,5 +129,5 @@ func (mcp *MCP) AutoDetect(filename, contentTypes string) (bool, int, error) {
// Schema returns the schema for the MCP converter
func (mcp *MCP) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "mcp", locale)
}

3
kb/providers/converters/ocr.go
View file

@ -7,6 +7,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/gou/pdf"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -169,5 +170,5 @@ func (ocr *OCR) AutoDetect(filename, contentTypes string) (bool, int, error) {
// Schema returns the schema for the OCR converter
func (ocr *OCR) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "ocr", locale)
}

3
kb/providers/converters/office.go
View file

@ -6,6 +6,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -109,5 +110,5 @@ func (office *Office) AutoDetect(filename, contentTypes string) (bool, int, erro
// Schema returns the schema for the Office converter
func (office *Office) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "office", locale)
}

3
kb/providers/converters/utf8.go
View file

@ -5,6 +5,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -44,5 +45,5 @@ func (utf8 *UTF8) AutoDetect(filename, contentTypes string) (bool, int, error) {
// Schema returns the schema for the UTF8 converter
func (utf8 *UTF8) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "utf8", locale)
}

3
kb/providers/converters/video.go
View file

@ -6,6 +6,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -190,5 +191,5 @@ func (video *Video) AutoDetect(filename, contentTypes string) (bool, int, error)
// Schema returns the schema for the Video converter
func (video *Video) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "video", locale)
}

3
kb/providers/converters/vision.go
View file

@ -5,6 +5,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -94,5 +95,5 @@ func (vision *Vision) AutoDetect(filename, contentTypes string) (bool, int, erro
// Schema returns the schema for the Vision converter
func (vision *Vision) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "vision", locale)
}

3
kb/providers/converters/whisper.go
View file

@ -5,6 +5,7 @@ import (
"github.com/yaoapp/gou/graphrag/converter"
"github.com/yaoapp/gou/graphrag/types"
"github.com/yaoapp/yao/kb/providers/factory"
kbtypes "github.com/yaoapp/yao/kb/types"
)
@ -144,5 +145,5 @@ func (whisper *Whisper) AutoDetect(filename, contentTypes string) (bool, int, er
// Schema returns the schema for the Whisper converter
func (whisper *Whisper) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeConverter, "whisper", locale)
}

4
kb/providers/embedding.go
View file

@ -71,7 +71,7 @@ func (o *OpenAI) Make(option *kbtypes.ProviderOption) (types.Embedding, error) {
// Schema returns the schema for the OpenAI embedding provider
func (o *OpenAI) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeEmbedding, "openai", locale)
}
// === Fastembed ===
@ -142,5 +142,5 @@ func (f *Fastembed) Make(option *kbtypes.ProviderOption) (types.Embedding, error
// Schema returns the schema for the Fastembed embedding provider
func (f *Fastembed) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeEmbedding, "fastembed", locale)
}

2
kb/providers/extractor.go
View file

@ -129,5 +129,5 @@ func (e *ExtractorOpenAI) Make(option *kbtypes.ProviderOption) (types.Extraction
// Schema returns the schema for the OpenAI extractor provider
func (e *ExtractorOpenAI) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeExtractor, "openai", locale)
}

4
kb/providers/fetcher.go
View file

@ -67,7 +67,7 @@ func (f *FetcherHTTP) Make(option *kbtypes.ProviderOption) (types.Fetcher, error
// Schema returns the schema for the HTTP fetcher provider
func (f *FetcherHTTP) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeFetcher, "http", locale)
}
// === FetcherMCP ===
@ -156,5 +156,5 @@ func (f *FetcherMCP) Make(option *kbtypes.ProviderOption) (types.Fetcher, error)
// Schema returns the schema for the MCP fetcher provider
func (f *FetcherMCP) Schema(provider *kbtypes.Provider, locale string) (*kbtypes.ProviderSchema, error) {
return nil, nil
return factory.GetSchemaFromBindata(factory.ProviderTypeFetcher, "mcp", locale)
}

1
yao/data/kb/providers/converter/en.json
View file

@ -1 +0,0 @@
{}

140
yao/data/kb/providers/converter/mcp/en.json Normal file
View file

@ -0,0 +1,140 @@
{
"id": "__yao.mcp",
"title": "MCP Tool Converter",
"description": "Model Context Protocol (MCP) converter that uses external tools for content conversion. Allows integration with custom conversion tools through the MCP interface.",
"required": ["id", "tool"],
"properties": {
"id": {
"type": "string",
"title": "MCP Server ID",
"description": "Identifier of the MCP server to use for conversion.",
"default": "",
"component": "Input",
"width": "half",
"order": 1
},
"tool": {
"type": "string",
"title": "Tool Name",
"description": "Name of the MCP tool to invoke for conversion.",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"arguments_mapping": {
"type": "object",
"title": "Arguments Mapping",
"description": "Mapping of conversion parameters to MCP tool arguments.",
"required": false,
"component": "Nested",
"width": "full",
"order": 3,
"properties": {
"file_path": {
"type": "string",
"title": "File Path Argument",
"description": "MCP tool argument name for the input file path.",
"default": "file_path",
"component": "Input",
"width": "half",
"order": 1
},
"content": {
"type": "string",
"title": "Content Argument",
"description": "MCP tool argument name for file content.",
"default": "content",
"component": "Input",
"width": "half",
"order": 2
},
"format": {
"type": "string",
"title": "Format Argument",
"description": "MCP tool argument name for output format specification.",
"default": "format",
"component": "Input",
"width": "half",
"order": 3
},
"options": {
"type": "string",
"title": "Options Argument",
"description": "MCP tool argument name for additional options.",
"default": "options",
"component": "Input",
"width": "half",
"order": 4
}
}
},
"result_mapping": {
"type": "object",
"title": "Result Mapping",
"description": "Mapping of MCP tool response fields to conversion results.",
"required": false,
"component": "Nested",
"width": "full",
"order": 4,
"properties": {
"content": {
"type": "string",
"title": "Content Field",
"description": "Response field containing the converted content.",
"default": "content",
"component": "Input",
"width": "half",
"order": 1
},
"metadata": {
"type": "string",
"title": "Metadata Field",
"description": "Response field containing conversion metadata.",
"default": "metadata",
"component": "Input",
"width": "half",
"order": 2
},
"error": {
"type": "string",
"title": "Error Field",
"description": "Response field containing error information.",
"default": "error",
"component": "Input",
"width": "half",
"order": 3
}
}
},
"notification_mapping": {
"type": "object",
"title": "Notification Mapping",
"description": "Mapping for MCP notification handling during conversion.",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"progress": {
"type": "string",
"title": "Progress Notification",
"description": "Notification type for conversion progress updates.",
"default": "progress",
"component": "Input",
"width": "half",
"order": 1
},
"status": {
"type": "string",
"title": "Status Notification",
"description": "Notification type for status changes.",
"default": "status",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

140
yao/data/kb/providers/converter/mcp/zh-cn.json Normal file
View file

@ -0,0 +1,140 @@
{
"id": "__yao.mcp",
"title": "MCP 工具转换器",
"description": "模型上下文协议（MCP）转换器，使用外部工具进行内容转换。允许通过 MCP 接口与自定义转换工具集成。",
"required": ["id", "tool"],
"properties": {
"id": {
"type": "string",
"title": "MCP 服务器 ID",
"description": "用于转换的 MCP 服务器标识符。",
"default": "",
"component": "Input",
"width": "half",
"order": 1
},
"tool": {
"type": "string",
"title": "工具名称",
"description": "用于转换的 MCP 工具名称。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"arguments_mapping": {
"type": "object",
"title": "参数映射",
"description": "转换参数到 MCP 工具参数的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 3,
"properties": {
"file_path": {
"type": "string",
"title": "文件路径参数",
"description": "输入文件路径的 MCP 工具参数名称。",
"default": "file_path",
"component": "Input",
"width": "half",
"order": 1
},
"content": {
"type": "string",
"title": "内容参数",
"description": "文件内容的 MCP 工具参数名称。",
"default": "content",
"component": "Input",
"width": "half",
"order": 2
},
"format": {
"type": "string",
"title": "格式参数",
"description": "输出格式规范的 MCP 工具参数名称。",
"default": "format",
"component": "Input",
"width": "half",
"order": 3
},
"options": {
"type": "string",
"title": "选项参数",
"description": "附加选项的 MCP 工具参数名称。",
"default": "options",
"component": "Input",
"width": "half",
"order": 4
}
}
},
"result_mapping": {
"type": "object",
"title": "结果映射",
"description": "MCP 工具响应字段到转换结果的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 4,
"properties": {
"content": {
"type": "string",
"title": "内容字段",
"description": "包含转换内容的响应字段。",
"default": "content",
"component": "Input",
"width": "half",
"order": 1
},
"metadata": {
"type": "string",
"title": "元数据字段",
"description": "包含转换元数据的响应字段。",
"default": "metadata",
"component": "Input",
"width": "half",
"order": 2
},
"error": {
"type": "string",
"title": "错误字段",
"description": "包含错误信息的响应字段。",
"default": "error",
"component": "Input",
"width": "half",
"order": 3
}
}
},
"notification_mapping": {
"type": "object",
"title": "通知映射",
"description": "转换期间 MCP 通知处理的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"progress": {
"type": "string",
"title": "进度通知",
"description": "转换进度更新的通知类型。",
"default": "progress",
"component": "Input",
"width": "half",
"order": 1
},
"status": {
"type": "string",
"title": "状态通知",
"description": "状态变更的通知类型。",
"default": "status",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

228
yao/data/kb/providers/converter/ocr/en.json Normal file
View file

@ -0,0 +1,228 @@
{
"id": "__yao.ocr",
"title": "OCR Text Recognition",
"description": "Optical Character Recognition for extracting text from images and PDF files. Supports various PDF conversion tools and configurable image processing options.",
"required": ["vision"],
"properties": {
"mode": {
"type": "string",
"title": "Processing Mode",
"description": "Processing mode for OCR operations.",
"default": "queue",
"enum": [
{
"label": "Queue Mode",
"value": "queue",
"description": "Process images sequentially in queue",
"default": true
},
{
"label": "Concurrent Mode",
"value": "concurrent",
"description": "Process images concurrently"
}
],
"component": "Select",
"width": "half",
"order": 1
},
"max_concurrency": {
"type": "integer",
"title": "Max Concurrency",
"description": "Maximum number of concurrent OCR processes.",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 2
},
"compress_size": {
"type": "integer",
"title": "Compress Size (KB)",
"description": "Maximum image size in KB before compression.",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 3
},
"force_image_mode": {
"type": "boolean",
"title": "Force Image Mode",
"description": "Force processing PDF as images instead of extracting text directly.",
"default": false,
"component": "Switch",
"width": "half",
"order": 4
},
"pdf_tool": {
"type": "string",
"title": "PDF Conversion Tool",
"description": "Tool used to convert PDF pages to images.",
"default": "pdftoppm",
"enum": [
{
"label": "pdftoppm",
"value": "pdftoppm",
"description": "Fast and reliable PDF to image conversion",
"default": true
},
{
"label": "mutool",
"value": "mutool",
"description": "MuPDF toolkit for PDF processing"
},
{
"label": "ImageMagick",
"value": "imagemagick",
"description": "Versatile image processing toolkit"
}
],
"component": "Select",
"width": "half",
"order": 5
},
"pdf_tool_path": {
"type": "string",
"title": "PDF Tool Path",
"description": "Custom path to the PDF conversion tool executable.",
"default": "",
"component": "Input",
"width": "half",
"order": 6
},
"pdf_dpi": {
"type": "integer",
"title": "PDF DPI",
"description": "Resolution in DPI when converting PDF to images.",
"default": 150,
"minimum": 72,
"maximum": 600,
"component": "InputNumber",
"width": "half",
"order": 7
},
"pdf_format": {
"type": "string",
"title": "PDF Image Format",
"description": "Image format for PDF conversion.",
"default": "png",
"enum": [
{
"label": "PNG",
"value": "png",
"description": "Lossless compression, best quality",
"default": true
},
{
"label": "JPEG",
"value": "jpeg",
"description": "Lossy compression, smaller file size"
}
],
"component": "Select",
"width": "half",
"order": 8
},
"pdf_quality": {
"type": "integer",
"title": "PDF JPEG Quality",
"description": "JPEG quality when using JPEG format (0-100).",
"default": 90,
"minimum": 10,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 9
},
"vision": {
"type": "object",
"title": "Vision Converter",
"description": "Vision AI converter configuration for text recognition.",
"required": true,
"component": "Nested",
"width": "full",
"order": 10,
"requiredFields": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "Vision Connector",
"description": "AI vision model connector for text recognition.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Vision",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "High-quality vision model with excellent OCR capabilities",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "Cost-effective vision model for basic OCR tasks"
}
]
},
{
"groupLabel": "Alternative Models",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "Excellent vision understanding and text extraction"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name",
"description": "Specific model name (optional, uses connector default).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "Custom Prompt",
"description": "Custom prompt for text extraction (optional).",
"default": "",
"component": "TextArea",
"width": "full",
"order": 3
},
"compress_size": {
"type": "integer",
"title": "Image Compress Size (KB)",
"description": "Maximum image size before compression for vision API.",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 4
},
"language": {
"type": "string",
"title": "Language",
"description": "Expected text language for better recognition.",
"default": "Auto",
"component": "Input",
"width": "half",
"order": 5
}
}
}
}
}

228
yao/data/kb/providers/converter/ocr/zh-cn.json Normal file
View file

@ -0,0 +1,228 @@
{
"id": "__yao.ocr",
"title": "OCR 文字识别",
"description": "光学字符识别，用于从图像和 PDF 文件中提取文本。支持多种 PDF 转换工具和可配置的图像处理选项。",
"required": ["vision"],
"properties": {
"mode": {
"type": "string",
"title": "处理模式",
"description": "OCR 操作的处理模式。",
"default": "queue",
"enum": [
{
"label": "队列模式",
"value": "queue",
"description": "按队列顺序处理图像",
"default": true
},
{
"label": "并发模式",
"value": "concurrent",
"description": "并发处理图像"
}
],
"component": "Select",
"width": "half",
"order": 1
},
"max_concurrency": {
"type": "integer",
"title": "最大并发数",
"description": "最大并发 OCR 进程数。",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 2
},
"compress_size": {
"type": "integer",
"title": "压缩大小 (KB)",
"description": "压缩前图像的最大大小（KB）。",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 3
},
"force_image_mode": {
"type": "boolean",
"title": "强制图像模式",
"description": "强制将 PDF 作为图像处理，而不是直接提取文本。",
"default": false,
"component": "Switch",
"width": "half",
"order": 4
},
"pdf_tool": {
"type": "string",
"title": "PDF 转换工具",
"description": "用于将 PDF 页面转换为图像的工具。",
"default": "pdftoppm",
"enum": [
{
"label": "pdftoppm",
"value": "pdftoppm",
"description": "快速可靠的 PDF 到图像转换",
"default": true
},
{
"label": "mutool",
"value": "mutool",
"description": "MuPDF 工具包用于 PDF 处理"
},
{
"label": "ImageMagick",
"value": "imagemagick",
"description": "多功能图像处理工具包"
}
],
"component": "Select",
"width": "half",
"order": 5
},
"pdf_tool_path": {
"type": "string",
"title": "PDF 工具路径",
"description": "PDF 转换工具可执行文件的自定义路径。",
"default": "",
"component": "Input",
"width": "half",
"order": 6
},
"pdf_dpi": {
"type": "integer",
"title": "PDF DPI",
"description": "将 PDF 转换为图像时的分辨率（DPI）。",
"default": 150,
"minimum": 72,
"maximum": 600,
"component": "InputNumber",
"width": "half",
"order": 7
},
"pdf_format": {
"type": "string",
"title": "PDF 图像格式",
"description": "PDF 转换的图像格式。",
"default": "png",
"enum": [
{
"label": "PNG",
"value": "png",
"description": "无损压缩，最佳质量",
"default": true
},
{
"label": "JPEG",
"value": "jpeg",
"description": "有损压缩，文件更小"
}
],
"component": "Select",
"width": "half",
"order": 8
},
"pdf_quality": {
"type": "integer",
"title": "PDF JPEG 质量",
"description": "使用 JPEG 格式时的质量 (0-100)。",
"default": 90,
"minimum": 10,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 9
},
"vision": {
"type": "object",
"title": "视觉转换器",
"description": "用于文本识别的视觉 AI 转换器配置。",
"required": true,
"component": "Nested",
"width": "full",
"order": 10,
"requiredFields": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "视觉连接器",
"description": "用于文本识别的 AI 视觉模型连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 视觉模型",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "高质量视觉模型，具有出色的 OCR 能力",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "基础 OCR 任务的经济型视觉模型"
}
]
},
{
"groupLabel": "其他模型",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "出色的视觉理解和文本提取能力"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称",
"description": "具体模型名称（可选，使用连接器默认值）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "自定义提示",
"description": "文本提取的自定义提示（可选）。",
"default": "",
"component": "TextArea",
"width": "full",
"order": 3
},
"compress_size": {
"type": "integer",
"title": "图像压缩大小 (KB)",
"description": "视觉 API 压缩前的最大图像大小。",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 4
},
"language": {
"type": "string",
"title": "语言",
"description": "预期文本语言，以便更好地识别。",
"default": "Auto",
"component": "Input",
"width": "half",
"order": 5
}
}
}
}
}

179
yao/data/kb/providers/converter/office/en.json Normal file
View file

@ -0,0 +1,179 @@
{
"id": "__yao.office",
"title": "Office Document Converter",
"description": "Converts Microsoft Office documents (DOCX, PPTX) to text. Supports extracting text, images, and multimedia content using AI vision and audio processing.",
"required": ["vision"],
"properties": {
"max_concurrency": {
"type": "integer",
"title": "Max Concurrency",
"description": "Maximum number of concurrent office processing operations.",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 1
},
"temp_dir": {
"type": "string",
"title": "Temporary Directory",
"description": "Custom temporary directory for file processing (empty = system temp).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"cleanup_temp": {
"type": "boolean",
"title": "Cleanup Temporary Files",
"description": "Automatically clean up temporary files after processing.",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"vision": {
"type": "object",
"title": "Vision Converter",
"description": "Vision AI converter for processing images and visual content in documents.",
"required": true,
"component": "Nested",
"width": "full",
"order": 4,
"requiredFields": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "Vision Connector",
"description": "AI vision model connector for image analysis.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Vision",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "High-quality vision model for detailed image analysis",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "Cost-effective vision model for basic image processing"
}
]
},
{
"groupLabel": "Alternative Models",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "Excellent vision understanding and content analysis"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name",
"description": "Specific model name (optional, uses connector default).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "Custom Prompt",
"description": "Custom prompt for image analysis (optional).",
"default": "",
"component": "TextArea",
"width": "full",
"order": 3
}
}
},
"video": {
"type": "object",
"title": "Video Converter (Optional)",
"description": "Video converter for processing video content in presentations.",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"keyframe_interval": {
"type": "number",
"title": "Keyframe Interval (seconds)",
"description": "Interval between extracted keyframes.",
"default": 10.0,
"minimum": 1.0,
"maximum": 60.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"max_keyframes": {
"type": "integer",
"title": "Max Keyframes",
"description": "Maximum number of keyframes to extract.",
"default": 20,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 2
}
}
},
"audio": {
"type": "object",
"title": "Audio Converter (Optional)",
"description": "Audio converter for processing audio content in presentations.",
"required": false,
"component": "Nested",
"width": "full",
"order": 6,
"properties": {
"connector": {
"type": "string",
"title": "Audio Connector",
"description": "AI audio transcription connector.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Audio",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "High-quality audio transcription",
"default": true
}
]
}
],
"width": "half",
"order": 1
},
"language": {
"type": "string",
"title": "Language",
"description": "Expected audio language (auto-detect if empty).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

179
yao/data/kb/providers/converter/office/zh-cn.json Normal file
View file

@ -0,0 +1,179 @@
{
"id": "__yao.office",
"title": "Office 文档转换器",
"description": "转换 Microsoft Office 文档（DOCX、PPTX）为文本。支持使用 AI 视觉和音频处理提取文本、图像和多媒体内容。",
"required": ["vision"],
"properties": {
"max_concurrency": {
"type": "integer",
"title": "最大并发数",
"description": "最大并发 Office 处理操作数。",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 1
},
"temp_dir": {
"type": "string",
"title": "临时目录",
"description": "文件处理的自定义临时目录（空 = 系统临时目录）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"cleanup_temp": {
"type": "boolean",
"title": "清理临时文件",
"description": "处理后自动清理临时文件。",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"vision": {
"type": "object",
"title": "视觉转换器",
"description": "用于处理文档中图像和视觉内容的视觉 AI 转换器。",
"required": true,
"component": "Nested",
"width": "full",
"order": 4,
"requiredFields": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "视觉连接器",
"description": "用于图像分析的 AI 视觉模型连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 视觉模型",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "用于详细图像分析的高质量视觉模型",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "基础图像处理的经济型视觉模型"
}
]
},
{
"groupLabel": "其他模型",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "出色的视觉理解和内容分析能力"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称",
"description": "具体模型名称（可选，使用连接器默认值）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "自定义提示",
"description": "图像分析的自定义提示（可选）。",
"default": "",
"component": "TextArea",
"width": "full",
"order": 3
}
}
},
"video": {
"type": "object",
"title": "视频转换器（可选）",
"description": "用于处理演示文稿中视频内容的视频转换器。",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"keyframe_interval": {
"type": "number",
"title": "关键帧间隔（秒）",
"description": "提取关键帧之间的间隔。",
"default": 10.0,
"minimum": 1.0,
"maximum": 60.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"max_keyframes": {
"type": "integer",
"title": "最大关键帧数",
"description": "要提取的最大关键帧数。",
"default": 20,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 2
}
}
},
"audio": {
"type": "object",
"title": "音频转换器（可选）",
"description": "用于处理演示文稿中音频内容的音频转换器。",
"required": false,
"component": "Nested",
"width": "full",
"order": 6,
"properties": {
"connector": {
"type": "string",
"title": "音频连接器",
"description": "AI 音频转录连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 音频",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "高质量音频转录",
"default": true
}
]
}
],
"width": "half",
"order": 1
},
"language": {
"type": "string",
"title": "语言",
"description": "预期音频语言（空则自动检测）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

7
yao/data/kb/providers/converter/utf8/en.json Normal file
View file

@ -0,0 +1,7 @@
{
"id": "__yao.utf8",
"title": "Plain Text Converter",
"description": "Processes plain text files (TXT, MD, etc.) by reading and normalizing UTF-8 content. This is the simplest converter that directly reads text files without any AI processing.",
"required": [],
"properties": {}
}

7
yao/data/kb/providers/converter/utf8/zh-cn.json Normal file
View file

@ -0,0 +1,7 @@
{
"id": "__yao.utf8",
"title": "纯文本转换器",
"description": "通过读取和规范化 UTF-8 内容来处理纯文本文件（TXT、MD 等）。这是最简单的转换器，直接读取文本文件而无需任何 AI 处理。",
"required": [],
"properties": {}
}

217
yao/data/kb/providers/converter/video/en.json Normal file
View file

@ -0,0 +1,217 @@
{
"id": "__yao.video",
"title": "Video Processing",
"description": "Processes video files by extracting keyframes and audio for AI analysis. Supports FFmpeg-based video processing with configurable GPU acceleration and quality settings.",
"required": [],
"properties": {
"keyframe_interval": {
"type": "number",
"title": "Keyframe Interval (seconds)",
"description": "Time interval between extracted keyframes for analysis.",
"default": 10.0,
"minimum": 1.0,
"maximum": 300.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"max_keyframes": {
"type": "integer",
"title": "Max Keyframes",
"description": "Maximum number of keyframes to extract from the video.",
"default": 20,
"minimum": 1,
"maximum": 1000,
"component": "InputNumber",
"width": "half",
"order": 2
},
"temp_dir": {
"type": "string",
"title": "Temporary Directory",
"description": "Custom temporary directory for video processing (empty = system temp).",
"default": "",
"component": "Input",
"width": "half",
"order": 3
},
"cleanup_temp": {
"type": "boolean",
"title": "Cleanup Temporary Files",
"description": "Automatically clean up temporary files after processing.",
"default": true,
"component": "Switch",
"width": "half",
"order": 4
},
"max_concurrency": {
"type": "integer",
"title": "Max Concurrency",
"description": "Maximum number of concurrent video processing operations.",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 5
},
"text_optimization": {
"type": "boolean",
"title": "Text Optimization",
"description": "Enable text optimization for better content extraction.",
"default": true,
"component": "Switch",
"width": "half",
"order": 6
},
"deduplication_ratio": {
"type": "number",
"title": "Deduplication Ratio",
"description": "Threshold for removing duplicate or similar keyframes (0.0-1.0).",
"default": 0.8,
"minimum": 0.0,
"maximum": 1.0,
"component": "InputNumber",
"width": "half",
"order": 7
},
"ffmpeg_path": {
"type": "string",
"title": "FFmpeg Path",
"description": "Custom path to FFmpeg executable (empty = use system PATH).",
"default": "",
"component": "Input",
"width": "half",
"order": 8
},
"ffprobe_path": {
"type": "string",
"title": "FFprobe Path",
"description": "Custom path to FFprobe executable (empty = use system PATH).",
"default": "",
"component": "Input",
"width": "half",
"order": 9
},
"enable_gpu": {
"type": "boolean",
"title": "Enable GPU Acceleration",
"description": "Use GPU acceleration for video processing if available.",
"default": false,
"component": "Switch",
"width": "half",
"order": 10
},
"gpu_index": {
"type": "integer",
"title": "GPU Index",
"description": "GPU device index to use (-1 = auto-detect, 0+ = specific GPU).",
"default": -1,
"minimum": -1,
"maximum": 7,
"component": "InputNumber",
"width": "half",
"order": 11
},
"max_threads": {
"type": "integer",
"title": "Max Threads",
"description": "Maximum number of threads for FFmpeg processing (-1 = auto).",
"default": -1,
"minimum": -1,
"maximum": 32,
"component": "InputNumber",
"width": "half",
"order": 12
},
"vision": {
"type": "object",
"title": "Vision Converter (Optional)",
"description": "Vision AI converter for analyzing extracted keyframes.",
"required": false,
"component": "Nested",
"width": "full",
"order": 13,
"properties": {
"connector": {
"type": "string",
"title": "Vision Connector",
"description": "AI vision model connector for keyframe analysis.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Vision",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "High-quality vision model for detailed frame analysis",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "Cost-effective vision model for basic frame analysis"
}
]
}
],
"width": "half",
"order": 1
},
"prompt": {
"type": "string",
"title": "Vision Prompt",
"description": "Custom prompt for keyframe analysis.",
"default": "",
"component": "TextArea",
"width": "full",
"order": 2
}
}
},
"audio": {
"type": "object",
"title": "Audio Converter (Optional)",
"description": "Audio converter for processing video soundtrack.",
"required": false,
"component": "Nested",
"width": "full",
"order": 14,
"properties": {
"connector": {
"type": "string",
"title": "Audio Connector",
"description": "AI audio transcription connector.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Audio",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "High-quality audio transcription",
"default": true
}
]
}
],
"width": "half",
"order": 1
},
"language": {
"type": "string",
"title": "Audio Language",
"description": "Expected audio language (empty = auto-detect).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

217
yao/data/kb/providers/converter/video/zh-cn.json Normal file
View file

@ -0,0 +1,217 @@
{
"id": "__yao.video",
"title": "视频处理",
"description": "通过提取关键帧和音频进行 AI 分析来处理视频文件。支持基于 FFmpeg 的视频处理，具有可配置的 GPU 加速和质量设置。",
"required": [],
"properties": {
"keyframe_interval": {
"type": "number",
"title": "关键帧间隔（秒）",
"description": "用于分析的提取关键帧之间的时间间隔。",
"default": 10.0,
"minimum": 1.0,
"maximum": 300.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"max_keyframes": {
"type": "integer",
"title": "最大关键帧数",
"description": "从视频中提取的最大关键帧数。",
"default": 20,
"minimum": 1,
"maximum": 1000,
"component": "InputNumber",
"width": "half",
"order": 2
},
"temp_dir": {
"type": "string",
"title": "临时目录",
"description": "视频处理的自定义临时目录（空 = 系统临时目录）。",
"default": "",
"component": "Input",
"width": "half",
"order": 3
},
"cleanup_temp": {
"type": "boolean",
"title": "清理临时文件",
"description": "处理后自动清理临时文件。",
"default": true,
"component": "Switch",
"width": "half",
"order": 4
},
"max_concurrency": {
"type": "integer",
"title": "最大并发数",
"description": "最大并发视频处理操作数。",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 5
},
"text_optimization": {
"type": "boolean",
"title": "文本优化",
"description": "启用文本优化以获得更好的内容提取。",
"default": true,
"component": "Switch",
"width": "half",
"order": 6
},
"deduplication_ratio": {
"type": "number",
"title": "去重比率",
"description": "移除重复或相似关键帧的阈值（0.0-1.0）。",
"default": 0.8,
"minimum": 0.0,
"maximum": 1.0,
"component": "InputNumber",
"width": "half",
"order": 7
},
"ffmpeg_path": {
"type": "string",
"title": "FFmpeg 路径",
"description": "FFmpeg 可执行文件的自定义路径（空 = 使用系统 PATH）。",
"default": "",
"component": "Input",
"width": "half",
"order": 8
},
"ffprobe_path": {
"type": "string",
"title": "FFprobe 路径",
"description": "FFprobe 可执行文件的自定义路径（空 = 使用系统 PATH）。",
"default": "",
"component": "Input",
"width": "half",
"order": 9
},
"enable_gpu": {
"type": "boolean",
"title": "启用 GPU 加速",
"description": "如果可用，使用 GPU 加速进行视频处理。",
"default": false,
"component": "Switch",
"width": "half",
"order": 10
},
"gpu_index": {
"type": "integer",
"title": "GPU 索引",
"description": "要使用的 GPU 设备索引（-1 = 自动检测，0+ = 特定 GPU）。",
"default": -1,
"minimum": -1,
"maximum": 7,
"component": "InputNumber",
"width": "half",
"order": 11
},
"max_threads": {
"type": "integer",
"title": "最大线程数",
"description": "FFmpeg 处理的最大线程数（-1 = 自动）。",
"default": -1,
"minimum": -1,
"maximum": 32,
"component": "InputNumber",
"width": "half",
"order": 12
},
"vision": {
"type": "object",
"title": "视觉转换器（可选）",
"description": "用于分析提取关键帧的视觉 AI 转换器。",
"required": false,
"component": "Nested",
"width": "full",
"order": 13,
"properties": {
"connector": {
"type": "string",
"title": "视觉连接器",
"description": "用于关键帧分析的 AI 视觉模型连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 视觉模型",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "用于详细帧分析的高质量视觉模型",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "基础帧分析的经济型视觉模型"
}
]
}
],
"width": "half",
"order": 1
},
"prompt": {
"type": "string",
"title": "视觉提示",
"description": "关键帧分析的自定义提示。",
"default": "",
"component": "TextArea",
"width": "full",
"order": 2
}
}
},
"audio": {
"type": "object",
"title": "音频转换器（可选）",
"description": "用于处理视频音轨的音频转换器。",
"required": false,
"component": "Nested",
"width": "full",
"order": 14,
"properties": {
"connector": {
"type": "string",
"title": "音频连接器",
"description": "AI 音频转录连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 音频",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "高质量音频转录",
"default": true
}
]
}
],
"width": "half",
"order": 1
},
"language": {
"type": "string",
"title": "音频语言",
"description": "预期音频语言（空 = 自动检测）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
}
}
}
}
}

136
yao/data/kb/providers/converter/vision/en.json Normal file
View file

@ -0,0 +1,136 @@
{
"id": "__yao.vision",
"title": "Vision Analysis",
"description": "Analyzes images using AI vision models to extract text, describe content, and identify objects. Supports various image formats with configurable compression and language settings.",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "Vision Connector",
"description": "AI vision model connector for image analysis.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Vision",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "High-quality vision model with detailed analysis capabilities",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "Cost-effective vision model for basic image analysis"
}
]
},
{
"groupLabel": "Alternative Models",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "Excellent vision understanding and detailed descriptions"
},
{
"label": "Google Gemini Vision",
"value": "google.gemini-vision",
"description": "Google's multimodal AI with vision capabilities"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name",
"description": "Specific model name (optional, uses connector default).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "Custom Prompt",
"description": "Custom prompt for image analysis. Leave empty to use default analysis prompt.",
"default": "",
"component": "TextArea",
"placeholder": "e.g., Describe this image in detail, focusing on text content and key visual elements.",
"width": "full",
"order": 3
},
"compress_size": {
"type": "integer",
"title": "Compress Size (KB)",
"description": "Maximum image size in KB before compression for API processing.",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 4
},
"language": {
"type": "string",
"title": "Language",
"description": "Expected content language for better analysis (Auto = automatic detection).",
"default": "Auto",
"component": "Select",
"enum": [
{
"label": "Auto Detect",
"value": "Auto",
"description": "Automatically detect content language",
"default": true
},
{
"label": "English",
"value": "English",
"description": "English content analysis"
},
{
"label": "Chinese",
"value": "Chinese",
"description": "Chinese content analysis"
},
{
"label": "Spanish",
"value": "Spanish",
"description": "Spanish content analysis"
},
{
"label": "French",
"value": "French",
"description": "French content analysis"
},
{
"label": "German",
"value": "German",
"description": "German content analysis"
},
{
"label": "Japanese",
"value": "Japanese",
"description": "Japanese content analysis"
}
],
"width": "half",
"order": 5
},
"options": {
"type": "string",
"title": "Additional Options (JSON)",
"description": "Additional model-specific options in JSON format.",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 6
}
}
}

136
yao/data/kb/providers/converter/vision/zh-cn.json Normal file
View file

@ -0,0 +1,136 @@
{
"id": "__yao.vision",
"title": "视觉分析",
"description": "使用 AI 视觉模型分析图像以提取文本、描述内容和识别对象。支持多种图像格式，具有可配置的压缩和语言设置。",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "视觉连接器",
"description": "用于图像分析的 AI 视觉模型连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 视觉模型",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "具有详细分析能力的高质量视觉模型",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "基础图像分析的经济型视觉模型"
}
]
},
{
"groupLabel": "其他模型",
"options": [
{
"label": "Claude 3.5 Sonnet",
"value": "anthropic.claude-3-5-sonnet",
"description": "出色的视觉理解和详细描述能力"
},
{
"label": "Google Gemini Vision",
"value": "google.gemini-vision",
"description": "Google 的多模态 AI，具有视觉能力"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称",
"description": "具体模型名称（可选，使用连接器默认值）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"prompt": {
"type": "string",
"title": "自定义提示",
"description": "图像分析的自定义提示。留空则使用默认分析提示。",
"default": "",
"component": "TextArea",
"placeholder": "例如：详细描述这张图像，重点关注文本内容和关键视觉元素。",
"width": "full",
"order": 3
},
"compress_size": {
"type": "integer",
"title": "压缩大小（KB）",
"description": "API 处理前压缩的最大图像大小（KB）。",
"default": 512,
"minimum": 100,
"maximum": 5120,
"component": "InputNumber",
"width": "half",
"order": 4
},
"language": {
"type": "string",
"title": "语言",
"description": "预期内容语言，以便更好地分析（自动 = 自动检测）。",
"default": "Auto",
"component": "Select",
"enum": [
{
"label": "自动检测",
"value": "Auto",
"description": "自动检测内容语言",
"default": true
},
{
"label": "英语",
"value": "English",
"description": "英语内容分析"
},
{
"label": "中文",
"value": "Chinese",
"description": "中文内容分析"
},
{
"label": "西班牙语",
"value": "Spanish",
"description": "西班牙语内容分析"
},
{
"label": "法语",
"value": "French",
"description": "法语内容分析"
},
{
"label": "德语",
"value": "German",
"description": "德语内容分析"
},
{
"label": "日语",
"value": "Japanese",
"description": "日语内容分析"
}
],
"width": "half",
"order": 5
},
"options": {
"type": "string",
"title": "附加选项（JSON）",
"description": "JSON 格式的附加模型特定选项。",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 6
}
}
}

155
yao/data/kb/providers/converter/whisper/en.json Normal file
View file

@ -0,0 +1,155 @@
{
"id": "__yao.whisper",
"title": "Audio Transcription",
"description": "Converts audio files to text using AI speech recognition. Supports automatic language detection, silence detection, and chunk-based processing for long audio files.",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "Audio Connector",
"description": "AI audio transcription connector.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Audio",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "High-quality multilingual audio transcription",
"default": true
}
]
},
{
"groupLabel": "Alternative Models",
"options": [
{
"label": "Azure Speech",
"value": "azure.speech",
"description": "Microsoft Azure Speech Services"
},
{
"label": "Google Speech",
"value": "google.speech",
"description": "Google Cloud Speech-to-Text"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name",
"description": "Specific model name (optional, uses connector default).",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"language": {
"type": "string",
"title": "Language",
"description": "Expected audio language (empty = auto-detect).",
"default": "",
"component": "Input",
"placeholder": "e.g., en, zh, es, fr",
"width": "half",
"order": 3
},
"chunk_duration": {
"type": "number",
"title": "Chunk Duration (seconds)",
"description": "Duration of each audio chunk for processing.",
"default": 30.0,
"minimum": 5.0,
"maximum": 300.0,
"component": "InputNumber",
"width": "half",
"order": 4
},
"mapping_duration": {
"type": "number",
"title": "Mapping Duration (seconds)",
"description": "Duration for timestamp mapping accuracy.",
"default": 5.0,
"minimum": 1.0,
"maximum": 30.0,
"component": "InputNumber",
"width": "half",
"order": 5
},
"silence_threshold": {
"type": "number",
"title": "Silence Threshold (dB)",
"description": "Audio level threshold for silence detection.",
"default": -40.0,
"minimum": -60.0,
"maximum": -10.0,
"component": "InputNumber",
"width": "half",
"order": 6
},
"silence_min_length": {
"type": "number",
"title": "Min Silence Length (seconds)",
"description": "Minimum silence duration to consider as break.",
"default": 1.0,
"minimum": 0.1,
"maximum": 10.0,
"component": "InputNumber",
"width": "half",
"order": 7
},
"enable_silence_detection": {
"type": "boolean",
"title": "Enable Silence Detection",
"description": "Use silence detection for better chunk boundaries.",
"default": true,
"component": "Switch",
"width": "half",
"order": 8
},
"max_concurrency": {
"type": "integer",
"title": "Max Concurrency",
"description": "Maximum concurrent transcription requests.",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 9
},
"temp_dir": {
"type": "string",
"title": "Temporary Directory",
"description": "Custom temporary directory for audio processing (empty = system temp).",
"default": "",
"component": "Input",
"width": "half",
"order": 10
},
"cleanup_temp": {
"type": "boolean",
"title": "Cleanup Temporary Files",
"description": "Automatically clean up temporary files after processing.",
"default": true,
"component": "Switch",
"width": "half",
"order": 11
},
"options": {
"type": "string",
"title": "Additional Options (JSON)",
"description": "Additional model-specific options in JSON format.",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 12
}
}
}

155
yao/data/kb/providers/converter/whisper/zh-cn.json Normal file
View file

@ -0,0 +1,155 @@
{
"id": "__yao.whisper",
"title": "音频转录",
"description": "使用 AI 语音识别将音频文件转换为文本。支持自动语言检测、静音检测和长音频文件的分块处理。",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "音频连接器",
"description": "AI 音频转录连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 音频",
"options": [
{
"label": "Whisper",
"value": "openai.whisper-1",
"description": "高质量多语言音频转录",
"default": true
}
]
},
{
"groupLabel": "其他模型",
"options": [
{
"label": "Azure 语音",
"value": "azure.speech",
"description": "Microsoft Azure 语音服务"
},
{
"label": "Google 语音",
"value": "google.speech",
"description": "Google Cloud 语音转文本"
}
]
}
],
"width": "half",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称",
"description": "具体模型名称（可选，使用连接器默认值）。",
"default": "",
"component": "Input",
"width": "half",
"order": 2
},
"language": {
"type": "string",
"title": "语言",
"description": "预期音频语言（空 = 自动检测）。",
"default": "",
"component": "Input",
"placeholder": "例如：en, zh, es, fr",
"width": "half",
"order": 3
},
"chunk_duration": {
"type": "number",
"title": "分块持续时间（秒）",
"description": "每个音频分块的处理持续时间。",
"default": 30.0,
"minimum": 5.0,
"maximum": 300.0,
"component": "InputNumber",
"width": "half",
"order": 4
},
"mapping_duration": {
"type": "number",
"title": "映射持续时间（秒）",
"description": "时间戳映射精度的持续时间。",
"default": 5.0,
"minimum": 1.0,
"maximum": 30.0,
"component": "InputNumber",
"width": "half",
"order": 5
},
"silence_threshold": {
"type": "number",
"title": "静音阈值（dB）",
"description": "静音检测的音频级别阈值。",
"default": -40.0,
"minimum": -60.0,
"maximum": -10.0,
"component": "InputNumber",
"width": "half",
"order": 6
},
"silence_min_length": {
"type": "number",
"title": "最小静音长度（秒）",
"description": "被视为中断的最小静音持续时间。",
"default": 1.0,
"minimum": 0.1,
"maximum": 10.0,
"component": "InputNumber",
"width": "half",
"order": 7
},
"enable_silence_detection": {
"type": "boolean",
"title": "启用静音检测",
"description": "使用静音检测获得更好的分块边界。",
"default": true,
"component": "Switch",
"width": "half",
"order": 8
},
"max_concurrency": {
"type": "integer",
"title": "最大并发数",
"description": "最大并发转录请求数。",
"default": 4,
"minimum": 1,
"maximum": 20,
"component": "InputNumber",
"width": "half",
"order": 9
},
"temp_dir": {
"type": "string",
"title": "临时目录",
"description": "音频处理的自定义临时目录（空 = 系统临时目录）。",
"default": "",
"component": "Input",
"width": "half",
"order": 10
},
"cleanup_temp": {
"type": "boolean",
"title": "清理临时文件",
"description": "处理后自动清理临时文件。",
"default": true,
"component": "Switch",
"width": "half",
"order": 11
},
"options": {
"type": "string",
"title": "附加选项（JSON）",
"description": "JSON 格式的附加模型特定选项。",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 12
}
}
}

1
yao/data/kb/providers/converter/zh-cn.json
View file

@ -1 +0,0 @@
{}

1
yao/data/kb/providers/embedding/en.json
View file

@ -1 +0,0 @@
{}

232
yao/data/kb/providers/embedding/fastembed/en.json Normal file
View file

@ -0,0 +1,232 @@
{
"id": "__yao.fastembed",
"title": "FastEmbed Embeddings",
"description": "FastEmbed local and remote embedding models for high-performance text vectorization. Supports various open-source models with configurable hosting options.",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "FastEmbed Connector",
"description": "FastEmbed connector for embedding model access.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "Local Models",
"options": [
{
"label": "BAAI/bge-small-en-v1.5",
"value": "fastembed.bge-small-en-v1.5",
"description": "384 dimensions, English optimized, fast inference",
"default": true
},
{
"label": "BAAI/bge-base-en-v1.5",
"value": "fastembed.bge-base-en-v1.5",
"description": "768 dimensions, English optimized, balanced performance"
},
{
"label": "BAAI/bge-large-en-v1.5",
"value": "fastembed.bge-large-en-v1.5",
"description": "1024 dimensions, English optimized, highest quality"
}
]
},
{
"groupLabel": "Multilingual Models",
"options": [
{
"label": "BAAI/bge-small-zh-v1.5",
"value": "fastembed.bge-small-zh-v1.5",
"description": "384 dimensions, Chinese optimized"
},
{
"label": "sentence-transformers/all-MiniLM-L6-v2",
"value": "fastembed.all-MiniLM-L6-v2",
"description": "384 dimensions, multilingual support"
},
{
"label": "sentence-transformers/all-mpnet-base-v2",
"value": "fastembed.all-mpnet-base-v2",
"description": "768 dimensions, multilingual, high quality"
}
]
},
{
"groupLabel": "Remote API",
"options": [
{
"label": "Custom API Endpoint",
"value": "fastembed.custom",
"description": "Connect to custom FastEmbed API endpoint"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name (Optional)",
"description": "Specific model name to override connector default.",
"default": "",
"component": "Input",
"placeholder": "e.g., BAAI/bge-small-en-v1.5",
"width": "half",
"order": 2
},
"dimensions": {
"type": "integer",
"title": "Embedding Dimensions",
"description": "Number of dimensions for the embedding vectors.",
"default": 384,
"component": "Select",
"enum": [
{
"label": "384 (BGE Small default)",
"value": 384,
"description": "Standard dimension for small models",
"default": true
},
{
"label": "768 (BGE Base default)",
"value": 768,
"description": "Standard dimension for base models"
},
{
"label": "1024 (BGE Large default)",
"value": 1024,
"description": "Standard dimension for large models"
},
{
"label": "512",
"value": 512,
"description": "Custom dimension for specific models"
},
{
"label": "256",
"value": 256,
"description": "Reduced dimension for memory efficiency"
}
],
"width": "half",
"order": 3
},
"concurrent": {
"type": "integer",
"title": "Concurrent Requests",
"description": "Number of concurrent embedding requests to process.",
"default": 10,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 4
},
"host": {
"type": "string",
"title": "API Host (Optional)",
"description": "Custom API host for remote FastEmbed service.",
"default": "",
"component": "Input",
"placeholder": "https://api.example.com",
"width": "half",
"order": 5
},
"key": {
"type": "string",
"title": "API Key (Optional)",
"description": "API key for authentication with remote service.",
"default": "",
"component": "Input",
"placeholder": "your-api-key-here",
"width": "half",
"order": 6
},
"batch_size": {
"type": "integer",
"title": "Batch Size",
"description": "Number of texts to embed in each batch.",
"default": 32,
"minimum": 1,
"maximum": 512,
"component": "InputNumber",
"width": "half",
"order": 7
},
"max_length": {
"type": "integer",
"title": "Max Token Length",
"description": "Maximum token length for input texts.",
"default": 512,
"minimum": 64,
"maximum": 8192,
"component": "InputNumber",
"width": "half",
"order": 8
},
"normalize": {
"type": "boolean",
"title": "Normalize Vectors",
"description": "Normalize embedding vectors to unit length.",
"default": true,
"component": "Switch",
"width": "half",
"order": 9
},
"cache_dir": {
"type": "string",
"title": "Model Cache Directory",
"description": "Directory to cache downloaded models (empty = default cache).",
"default": "",
"component": "Input",
"placeholder": "/path/to/model/cache",
"width": "half",
"order": 10
},
"device": {
"type": "string",
"title": "Compute Device",
"description": "Device to use for model inference.",
"default": "cpu",
"component": "Select",
"enum": [
{
"label": "CPU",
"value": "cpu",
"description": "Use CPU for inference (compatible with all systems)",
"default": true
},
{
"label": "CUDA (GPU)",
"value": "cuda",
"description": "Use NVIDIA GPU for faster inference"
},
{
"label": "MPS (Apple Silicon)",
"value": "mps",
"description": "Use Apple Silicon GPU acceleration"
},
{
"label": "Auto",
"value": "auto",
"description": "Automatically select best available device"
}
],
"width": "half",
"order": 11
},
"threads": {
"type": "integer",
"title": "CPU Threads",
"description": "Number of CPU threads to use (-1 = auto).",
"default": -1,
"minimum": -1,
"maximum": 32,
"component": "InputNumber",
"width": "half",
"order": 12
}
}
}

232
yao/data/kb/providers/embedding/fastembed/zh-cn.json Normal file
View file

@ -0,0 +1,232 @@
{
"id": "__yao.fastembed",
"title": "FastEmbed 嵌入模型",
"description": "FastEmbed 本地和远程嵌入模型，用于高性能文本向量化。支持各种开源模型和可配置的托管选项。",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "FastEmbed 连接器",
"description": "用于嵌入模型访问的 FastEmbed 连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "本地模型",
"options": [
{
"label": "BAAI/bge-small-en-v1.5",
"value": "fastembed.bge-small-en-v1.5",
"description": "384 维度，英语优化，快速推理",
"default": true
},
{
"label": "BAAI/bge-base-en-v1.5",
"value": "fastembed.bge-base-en-v1.5",
"description": "768 维度，英语优化，平衡性能"
},
{
"label": "BAAI/bge-large-en-v1.5",
"value": "fastembed.bge-large-en-v1.5",
"description": "1024 维度，英语优化，最高质量"
}
]
},
{
"groupLabel": "多语言模型",
"options": [
{
"label": "BAAI/bge-small-zh-v1.5",
"value": "fastembed.bge-small-zh-v1.5",
"description": "384 维度，中文优化"
},
{
"label": "sentence-transformers/all-MiniLM-L6-v2",
"value": "fastembed.all-MiniLM-L6-v2",
"description": "384 维度，多语言支持"
},
{
"label": "sentence-transformers/all-mpnet-base-v2",
"value": "fastembed.all-mpnet-base-v2",
"description": "768 维度，多语言，高质量"
}
]
},
{
"groupLabel": "远程 API",
"options": [
{
"label": "自定义 API 端点",
"value": "fastembed.custom",
"description": "连接到自定义 FastEmbed API 端点"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称（可选）",
"description": "覆盖连接器默认值的特定模型名称。",
"default": "",
"component": "Input",
"placeholder": "例如：BAAI/bge-small-en-v1.5",
"width": "half",
"order": 2
},
"dimensions": {
"type": "integer",
"title": "嵌入维度",
"description": "嵌入向量的维度数量。",
"default": 384,
"component": "Select",
"enum": [
{
"label": "384 (BGE Small 默认)",
"value": 384,
"description": "小型模型的标准维度",
"default": true
},
{
"label": "768 (BGE Base 默认)",
"value": 768,
"description": "基础模型的标准维度"
},
{
"label": "1024 (BGE Large 默认)",
"value": 1024,
"description": "大型模型的标准维度"
},
{
"label": "512",
"value": 512,
"description": "特定模型的自定义维度"
},
{
"label": "256",
"value": 256,
"description": "降低维度以提高内存效率"
}
],
"width": "half",
"order": 3
},
"concurrent": {
"type": "integer",
"title": "并发请求数",
"description": "要处理的并发嵌入请求数量。",
"default": 10,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 4
},
"host": {
"type": "string",
"title": "API 主机（可选）",
"description": "远程 FastEmbed 服务的自定义 API 主机。",
"default": "",
"component": "Input",
"placeholder": "https://api.example.com",
"width": "half",
"order": 5
},
"key": {
"type": "string",
"title": "API 密钥（可选）",
"description": "用于远程服务身份验证的 API 密钥。",
"default": "",
"component": "Input",
"placeholder": "your-api-key-here",
"width": "half",
"order": 6
},
"batch_size": {
"type": "integer",
"title": "批处理大小",
"description": "每批嵌入的文本数量。",
"default": 32,
"minimum": 1,
"maximum": 512,
"component": "InputNumber",
"width": "half",
"order": 7
},
"max_length": {
"type": "integer",
"title": "最大令牌长度",
"description": "输入文本的最大令牌长度。",
"default": 512,
"minimum": 64,
"maximum": 8192,
"component": "InputNumber",
"width": "half",
"order": 8
},
"normalize": {
"type": "boolean",
"title": "规范化向量",
"description": "将嵌入向量规范化为单位长度。",
"default": true,
"component": "Switch",
"width": "half",
"order": 9
},
"cache_dir": {
"type": "string",
"title": "模型缓存目录",
"description": "缓存下载模型的目录（空 = 默认缓存）。",
"default": "",
"component": "Input",
"placeholder": "/path/to/model/cache",
"width": "half",
"order": 10
},
"device": {
"type": "string",
"title": "计算设备",
"description": "用于模型推理的设备。",
"default": "cpu",
"component": "Select",
"enum": [
{
"label": "CPU",
"value": "cpu",
"description": "使用 CPU 进行推理（兼容所有系统）",
"default": true
},
{
"label": "CUDA (GPU)",
"value": "cuda",
"description": "使用 NVIDIA GPU 进行更快推理"
},
{
"label": "MPS (Apple Silicon)",
"value": "mps",
"description": "使用 Apple Silicon GPU 加速"
},
{
"label": "自动",
"value": "auto",
"description": "自动选择最佳可用设备"
}
],
"width": "half",
"order": 11
},
"threads": {
"type": "integer",
"title": "CPU 线程数",
"description": "要使用的 CPU 线程数（-1 = 自动）。",
"default": -1,
"minimum": -1,
"maximum": 32,
"component": "InputNumber",
"width": "half",
"order": 12
}
}
}

186
yao/data/kb/providers/embedding/openai/en.json Normal file
View file

@ -0,0 +1,186 @@
{
"id": "__yao.openai",
"title": "OpenAI Embeddings",
"description": "OpenAI's text embedding models for converting text into high-dimensional vectors. Supports various embedding models with configurable dimensions and concurrent processing.",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "OpenAI Connector",
"description": "OpenAI connector for embedding API access.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Models",
"options": [
{
"label": "text-embedding-3-small",
"value": "openai.text-embedding-3-small",
"description": "Latest small embedding model, 1536 dimensions, cost-effective",
"default": true
},
{
"label": "text-embedding-3-large",
"value": "openai.text-embedding-3-large",
"description": "Latest large embedding model, 3072 dimensions, highest quality"
},
{
"label": "text-embedding-ada-002",
"value": "openai.text-embedding-ada-002",
"description": "Previous generation model, 1536 dimensions, reliable"
}
]
},
{
"groupLabel": "Azure OpenAI",
"options": [
{
"label": "Azure text-embedding-3-small",
"value": "azure.text-embedding-3-small",
"description": "Azure hosted embedding model, 1536 dimensions"
},
{
"label": "Azure text-embedding-3-large",
"value": "azure.text-embedding-3-large",
"description": "Azure hosted large embedding model, 3072 dimensions"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name (Optional)",
"description": "Specific model name to override connector default.",
"default": "",
"component": "Select",
"enum": [
{
"label": "Use Connector Default",
"value": "",
"description": "Use the default model from the selected connector",
"default": true
},
{
"label": "text-embedding-3-small",
"value": "text-embedding-3-small",
"description": "1536 dimensions, cost-effective"
},
{
"label": "text-embedding-3-large",
"value": "text-embedding-3-large",
"description": "3072 dimensions, highest quality"
},
{
"label": "text-embedding-ada-002",
"value": "text-embedding-ada-002",
"description": "1536 dimensions, previous generation"
}
],
"width": "half",
"order": 2
},
"dimensions": {
"type": "integer",
"title": "Embedding Dimensions",
"description": "Number of dimensions for the embedding vectors.",
"default": 1536,
"component": "Select",
"enum": [
{
"label": "1536 (text-embedding-3-small default)",
"value": 1536,
"description": "Standard dimension for text-embedding-3-small",
"default": true
},
{
"label": "3072 (text-embedding-3-large default)",
"value": 3072,
"description": "Full dimension for text-embedding-3-large"
},
{
"label": "1024",
"value": 1024,
"description": "Reduced dimension for smaller storage"
},
{
"label": "512",
"value": 512,
"description": "Compact dimension for memory efficiency"
},
{
"label": "256",
"value": 256,
"description": "Minimal dimension for basic similarity"
}
],
"width": "half",
"order": 3
},
"concurrent": {
"type": "integer",
"title": "Concurrent Requests",
"description": "Number of concurrent embedding requests to process.",
"default": 10,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 4
},
"batch_size": {
"type": "integer",
"title": "Batch Size",
"description": "Number of texts to embed in each API request.",
"default": 100,
"minimum": 1,
"maximum": 2048,
"component": "InputNumber",
"width": "half",
"order": 5
},
"retry_attempts": {
"type": "integer",
"title": "Retry Attempts",
"description": "Number of retry attempts for failed requests.",
"default": 3,
"minimum": 0,
"maximum": 10,
"component": "InputNumber",
"width": "half",
"order": 6
},
"timeout": {
"type": "integer",
"title": "Timeout (seconds)",
"description": "Request timeout in seconds.",
"default": 30,
"minimum": 5,
"maximum": 300,
"component": "InputNumber",
"width": "half",
"order": 7
},
"normalize": {
"type": "boolean",
"title": "Normalize Vectors",
"description": "Normalize embedding vectors to unit length.",
"default": true,
"component": "Switch",
"width": "half",
"order": 8
},
"truncate_input": {
"type": "boolean",
"title": "Truncate Long Input",
"description": "Automatically truncate input text that exceeds token limits.",
"default": true,
"component": "Switch",
"width": "half",
"order": 9
}
}
}

186
yao/data/kb/providers/embedding/openai/zh-cn.json Normal file
View file

@ -0,0 +1,186 @@
{
"id": "__yao.openai",
"title": "OpenAI 嵌入模型",
"description": "OpenAI 的文本嵌入模型，用于将文本转换为高维向量。支持各种嵌入模型，具有可配置的维度和并发处理。",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "OpenAI 连接器",
"description": "用于嵌入 API 访问的 OpenAI 连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 模型",
"options": [
{
"label": "text-embedding-3-small",
"value": "openai.text-embedding-3-small",
"description": "最新的小型嵌入模型，1536 维度，经济高效",
"default": true
},
{
"label": "text-embedding-3-large",
"value": "openai.text-embedding-3-large",
"description": "最新的大型嵌入模型，3072 维度，最高质量"
},
{
"label": "text-embedding-ada-002",
"value": "openai.text-embedding-ada-002",
"description": "上一代模型，1536 维度，可靠稳定"
}
]
},
{
"groupLabel": "Azure OpenAI",
"options": [
{
"label": "Azure text-embedding-3-small",
"value": "azure.text-embedding-3-small",
"description": "Azure 托管的嵌入模型，1536 维度"
},
{
"label": "Azure text-embedding-3-large",
"value": "azure.text-embedding-3-large",
"description": "Azure 托管的大型嵌入模型，3072 维度"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称（可选）",
"description": "覆盖连接器默认值的特定模型名称。",
"default": "",
"component": "Select",
"enum": [
{
"label": "使用连接器默认",
"value": "",
"description": "使用所选连接器的默认模型",
"default": true
},
{
"label": "text-embedding-3-small",
"value": "text-embedding-3-small",
"description": "1536 维度，经济高效"
},
{
"label": "text-embedding-3-large",
"value": "text-embedding-3-large",
"description": "3072 维度，最高质量"
},
{
"label": "text-embedding-ada-002",
"value": "text-embedding-ada-002",
"description": "1536 维度，上一代"
}
],
"width": "half",
"order": 2
},
"dimensions": {
"type": "integer",
"title": "嵌入维度",
"description": "嵌入向量的维度数量。",
"default": 1536,
"component": "Select",
"enum": [
{
"label": "1536 (text-embedding-3-small 默认)",
"value": 1536,
"description": "text-embedding-3-small 的标准维度",
"default": true
},
{
"label": "3072 (text-embedding-3-large 默认)",
"value": 3072,
"description": "text-embedding-3-large 的完整维度"
},
{
"label": "1024",
"value": 1024,
"description": "降低维度以减少存储空间"
},
{
"label": "512",
"value": 512,
"description": "紧凑维度以提高内存效率"
},
{
"label": "256",
"value": 256,
"description": "最小维度用于基本相似性"
}
],
"width": "half",
"order": 3
},
"concurrent": {
"type": "integer",
"title": "并发请求数",
"description": "要处理的并发嵌入请求数量。",
"default": 10,
"minimum": 1,
"maximum": 100,
"component": "InputNumber",
"width": "half",
"order": 4
},
"batch_size": {
"type": "integer",
"title": "批处理大小",
"description": "每个 API 请求中嵌入的文本数量。",
"default": 100,
"minimum": 1,
"maximum": 2048,
"component": "InputNumber",
"width": "half",
"order": 5
},
"retry_attempts": {
"type": "integer",
"title": "重试次数",
"description": "失败请求的重试次数。",
"default": 3,
"minimum": 0,
"maximum": 10,
"component": "InputNumber",
"width": "half",
"order": 6
},
"timeout": {
"type": "integer",
"title": "超时时间（秒）",
"description": "请求超时时间，以秒为单位。",
"default": 30,
"minimum": 5,
"maximum": 300,
"component": "InputNumber",
"width": "half",
"order": 7
},
"normalize": {
"type": "boolean",
"title": "规范化向量",
"description": "将嵌入向量规范化为单位长度。",
"default": true,
"component": "Switch",
"width": "half",
"order": 8
},
"truncate_input": {
"type": "boolean",
"title": "截断长输入",
"description": "自动截断超过令牌限制的输入文本。",
"default": true,
"component": "Switch",
"width": "half",
"order": 9
}
}
}

1
yao/data/kb/providers/embedding/zh-cn.json
View file

@ -1 +0,0 @@
{}

1
yao/data/kb/providers/extractor/en.json
View file

@ -1 +0,0 @@
{}

296
yao/data/kb/providers/extractor/openai/en.json Normal file
View file

@ -0,0 +1,296 @@
{
"id": "__yao.openai",
"title": "OpenAI Entity Extractor",
"description": "AI-powered entity and relationship extraction using OpenAI's language models. Extracts structured information from unstructured text with configurable prompts and tool calling capabilities.",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "OpenAI Connector",
"description": "OpenAI connector for extraction API access.",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI Models",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "Latest GPT-4o model, excellent for complex extraction tasks",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "Cost-effective GPT-4o Mini for basic extraction"
},
{
"label": "GPT-4 Turbo",
"value": "openai.gpt-4-turbo",
"description": "GPT-4 Turbo with enhanced performance"
},
{
"label": "GPT-3.5 Turbo",
"value": "openai.gpt-3.5-turbo",
"description": "Fast and economical for simple extractions"
}
]
},
{
"groupLabel": "Azure OpenAI",
"options": [
{
"label": "Azure GPT-4o",
"value": "azure.gpt-4o",
"description": "Azure hosted GPT-4o model"
},
{
"label": "Azure GPT-4",
"value": "azure.gpt-4",
"description": "Azure hosted GPT-4 model"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "Model Name (Optional)",
"description": "Specific model name to override connector default.",
"default": "",
"component": "Select",
"enum": [
{
"label": "Use Connector Default",
"value": "",
"description": "Use the default model from the selected connector",
"default": true
},
{
"label": "gpt-4o",
"value": "gpt-4o",
"description": "Latest GPT-4o model"
},
{
"label": "gpt-4o-mini",
"value": "gpt-4o-mini",
"description": "Cost-effective GPT-4o Mini"
},
{
"label": "gpt-4-turbo",
"value": "gpt-4-turbo",
"description": "GPT-4 Turbo model"
},
{
"label": "gpt-3.5-turbo",
"value": "gpt-3.5-turbo",
"description": "GPT-3.5 Turbo model"
}
],
"width": "half",
"order": 2
},
"toolcall": {
"type": "boolean",
"title": "Enable Tool Calling",
"description": "Use structured tool calling for more reliable extraction.",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"temperature": {
"type": "number",
"title": "Temperature",
"description": "Controls randomness in extraction (0.0 = deterministic, 1.0 = creative).",
"default": 0.1,
"minimum": 0.0,
"maximum": 2.0,
"component": "InputNumber",
"width": "half",
"order": 4
},
"max_tokens": {
"type": "integer",
"title": "Max Tokens",
"description": "Maximum number of tokens in the response.",
"default": 4000,
"minimum": 100,
"maximum": 32000,
"component": "InputNumber",
"width": "half",
"order": 5
},
"concurrent": {
"type": "integer",
"title": "Concurrent Requests",
"description": "Number of concurrent extraction requests to process.",
"default": 5,
"minimum": 1,
"maximum": 50,
"component": "InputNumber",
"width": "half",
"order": 6
},
"retry_attempts": {
"type": "integer",
"title": "Retry Attempts",
"description": "Number of retry attempts for failed extractions.",
"default": 3,
"minimum": 0,
"maximum": 10,
"component": "InputNumber",
"width": "half",
"order": 7
},
"retry_delay": {
"type": "number",
"title": "Retry Delay (seconds)",
"description": "Delay between retry attempts in seconds.",
"default": 1.0,
"minimum": 0.1,
"maximum": 60.0,
"component": "InputNumber",
"width": "half",
"order": 8
},
"prompt": {
"type": "string",
"title": "Custom Extraction Prompt",
"description": "Custom prompt for entity and relationship extraction (optional).",
"default": "",
"component": "TextArea",
"placeholder": "You are an expert at extracting entities and relationships from text. Extract all relevant entities and their relationships in a structured format...",
"width": "full",
"order": 9
},
"extraction_types": {
"type": "object",
"title": "Extraction Configuration",
"description": "Configure what types of entities and relationships to extract.",
"required": false,
"component": "Nested",
"width": "full",
"order": 10,
"properties": {
"entities": {
"type": "boolean",
"title": "Extract Entities",
"description": "Extract named entities (people, places, organizations, etc.).",
"default": true,
"component": "Switch",
"width": "half",
"order": 1
},
"relationships": {
"type": "boolean",
"title": "Extract Relationships",
"description": "Extract relationships between entities.",
"default": true,
"component": "Switch",
"width": "half",
"order": 2
},
"concepts": {
"type": "boolean",
"title": "Extract Concepts",
"description": "Extract key concepts and themes.",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"summaries": {
"type": "boolean",
"title": "Generate Summaries",
"description": "Generate document summaries.",
"default": false,
"component": "Switch",
"width": "half",
"order": 4
},
"entity_types": {
"type": "string",
"title": "Entity Types",
"description": "Comma-separated list of entity types to focus on (optional).",
"default": "",
"component": "Input",
"placeholder": "PERSON, ORGANIZATION, LOCATION, EVENT, PRODUCT",
"width": "full",
"order": 5
},
"relationship_types": {
"type": "string",
"title": "Relationship Types",
"description": "Comma-separated list of relationship types to focus on (optional).",
"default": "",
"component": "Input",
"placeholder": "WORKS_FOR, LOCATED_IN, PART_OF, RELATED_TO",
"width": "full",
"order": 6
}
}
},
"tools": {
"type": "string",
"title": "Custom Tools (JSON)",
"description": "Custom tool definitions for structured extraction (advanced usage).",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 11
},
"quality_settings": {
"type": "object",
"title": "Quality Settings",
"description": "Fine-tune extraction quality and performance.",
"required": false,
"component": "Nested",
"width": "full",
"order": 12,
"properties": {
"confidence_threshold": {
"type": "number",
"title": "Confidence Threshold",
"description": "Minimum confidence score for extracted entities (0.0-1.0).",
"default": 0.7,
"minimum": 0.0,
"maximum": 1.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"deduplicate": {
"type": "boolean",
"title": "Deduplicate Results",
"description": "Remove duplicate entities and relationships.",
"default": true,
"component": "Switch",
"width": "half",
"order": 2
},
"normalize_entities": {
"type": "boolean",
"title": "Normalize Entity Names",
"description": "Normalize entity names for consistency.",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"include_context": {
"type": "boolean",
"title": "Include Context",
"description": "Include surrounding context for extracted entities.",
"default": true,
"component": "Switch",
"width": "half",
"order": 4
}
}
}
}
}

296
yao/data/kb/providers/extractor/openai/zh-cn.json Normal file
View file

@ -0,0 +1,296 @@
{
"id": "__yao.openai",
"title": "OpenAI 实体提取器",
"description": "使用 OpenAI 语言模型进行 AI 驱动的实体和关系提取。从非结构化文本中提取结构化信息，支持可配置的提示和工具调用功能。",
"required": ["connector"],
"properties": {
"connector": {
"type": "string",
"title": "OpenAI 连接器",
"description": "用于提取 API 访问的 OpenAI 连接器。",
"default": "",
"component": "Select",
"enum": [
{
"groupLabel": "OpenAI 模型",
"options": [
{
"label": "GPT-4o",
"value": "openai.gpt-4o",
"description": "最新的 GPT-4o 模型，适用于复杂的提取任务",
"default": true
},
{
"label": "GPT-4o Mini",
"value": "openai.gpt-4o-mini",
"description": "经济高效的 GPT-4o Mini，适用于基础提取"
},
{
"label": "GPT-4 Turbo",
"value": "openai.gpt-4-turbo",
"description": "性能增强的 GPT-4 Turbo"
},
{
"label": "GPT-3.5 Turbo",
"value": "openai.gpt-3.5-turbo",
"description": "快速经济，适用于简单提取"
}
]
},
{
"groupLabel": "Azure OpenAI",
"options": [
{
"label": "Azure GPT-4o",
"value": "azure.gpt-4o",
"description": "Azure 托管的 GPT-4o 模型"
},
{
"label": "Azure GPT-4",
"value": "azure.gpt-4",
"description": "Azure 托管的 GPT-4 模型"
}
]
}
],
"width": "full",
"order": 1
},
"model": {
"type": "string",
"title": "模型名称（可选）",
"description": "覆盖连接器默认值的特定模型名称。",
"default": "",
"component": "Select",
"enum": [
{
"label": "使用连接器默认",
"value": "",
"description": "使用所选连接器的默认模型",
"default": true
},
{
"label": "gpt-4o",
"value": "gpt-4o",
"description": "最新的 GPT-4o 模型"
},
{
"label": "gpt-4o-mini",
"value": "gpt-4o-mini",
"description": "经济高效的 GPT-4o Mini"
},
{
"label": "gpt-4-turbo",
"value": "gpt-4-turbo",
"description": "GPT-4 Turbo 模型"
},
{
"label": "gpt-3.5-turbo",
"value": "gpt-3.5-turbo",
"description": "GPT-3.5 Turbo 模型"
}
],
"width": "half",
"order": 2
},
"toolcall": {
"type": "boolean",
"title": "启用工具调用",
"description": "使用结构化工具调用以获得更可靠的提取。",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"temperature": {
"type": "number",
"title": "温度",
"description": "控制提取的随机性（0.0 = 确定性，1.0 = 创造性）。",
"default": 0.1,
"minimum": 0.0,
"maximum": 2.0,
"component": "InputNumber",
"width": "half",
"order": 4
},
"max_tokens": {
"type": "integer",
"title": "最大令牌数",
"description": "响应中的最大令牌数。",
"default": 4000,
"minimum": 100,
"maximum": 32000,
"component": "InputNumber",
"width": "half",
"order": 5
},
"concurrent": {
"type": "integer",
"title": "并发请求数",
"description": "要处理的并发提取请求数。",
"default": 5,
"minimum": 1,
"maximum": 50,
"component": "InputNumber",
"width": "half",
"order": 6
},
"retry_attempts": {
"type": "integer",
"title": "重试次数",
"description": "失败提取的重试次数。",
"default": 3,
"minimum": 0,
"maximum": 10,
"component": "InputNumber",
"width": "half",
"order": 7
},
"retry_delay": {
"type": "number",
"title": "重试延迟（秒）",
"description": "重试尝试之间的延迟时间（秒）。",
"default": 1.0,
"minimum": 0.1,
"maximum": 60.0,
"component": "InputNumber",
"width": "half",
"order": 8
},
"prompt": {
"type": "string",
"title": "自定义提取提示",
"description": "用于实体和关系提取的自定义提示（可选）。",
"default": "",
"component": "TextArea",
"placeholder": "您是从文本中提取实体和关系的专家。请以结构化格式提取所有相关实体及其关系...",
"width": "full",
"order": 9
},
"extraction_types": {
"type": "object",
"title": "提取配置",
"description": "配置要提取的实体和关系类型。",
"required": false,
"component": "Nested",
"width": "full",
"order": 10,
"properties": {
"entities": {
"type": "boolean",
"title": "提取实体",
"description": "提取命名实体（人物、地点、组织等）。",
"default": true,
"component": "Switch",
"width": "half",
"order": 1
},
"relationships": {
"type": "boolean",
"title": "提取关系",
"description": "提取实体之间的关系。",
"default": true,
"component": "Switch",
"width": "half",
"order": 2
},
"concepts": {
"type": "boolean",
"title": "提取概念",
"description": "提取关键概念和主题。",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"summaries": {
"type": "boolean",
"title": "生成摘要",
"description": "生成文档摘要。",
"default": false,
"component": "Switch",
"width": "half",
"order": 4
},
"entity_types": {
"type": "string",
"title": "实体类型",
"description": "要重点关注的实体类型，用逗号分隔（可选）。",
"default": "",
"component": "Input",
"placeholder": "人物, 组织, 地点, 事件, 产品",
"width": "full",
"order": 5
},
"relationship_types": {
"type": "string",
"title": "关系类型",
"description": "要重点关注的关系类型，用逗号分隔（可选）。",
"default": "",
"component": "Input",
"placeholder": "工作于, 位于, 属于, 相关于",
"width": "full",
"order": 6
}
}
},
"tools": {
"type": "string",
"title": "自定义工具（JSON）",
"description": "用于结构化提取的自定义工具定义（高级用法）。",
"default": "",
"component": "CodeEditor",
"width": "full",
"order": 11
},
"quality_settings": {
"type": "object",
"title": "质量设置",
"description": "微调提取质量和性能。",
"required": false,
"component": "Nested",
"width": "full",
"order": 12,
"properties": {
"confidence_threshold": {
"type": "number",
"title": "置信度阈值",
"description": "提取实体的最小置信度分数（0.0-1.0）。",
"default": 0.7,
"minimum": 0.0,
"maximum": 1.0,
"component": "InputNumber",
"width": "half",
"order": 1
},
"deduplicate": {
"type": "boolean",
"title": "去重结果",
"description": "移除重复的实体和关系。",
"default": true,
"component": "Switch",
"width": "half",
"order": 2
},
"normalize_entities": {
"type": "boolean",
"title": "规范化实体名称",
"description": "规范化实体名称以保持一致性。",
"default": true,
"component": "Switch",
"width": "half",
"order": 3
},
"include_context": {
"type": "boolean",
"title": "包含上下文",
"description": "为提取的实体包含周围上下文。",
"default": true,
"component": "Switch",
"width": "half",
"order": 4
}
}
}
}
}

1
yao/data/kb/providers/extractor/zh-cn.json
View file

@ -1 +0,0 @@
{}

0
yao/data/kb/providers/fetcher/en.json
View file

188
yao/data/kb/providers/fetcher/http/en.json Normal file
View file

@ -0,0 +1,188 @@
{
"id": "__yao.http",
"title": "HTTP Web Fetcher",
"description": "Fetches content from web URLs using HTTP/HTTPS protocols. Supports custom headers, user agent configuration, and timeout settings for reliable web content retrieval.",
"required": [],
"properties": {
"headers": {
"type": "object",
"title": "Custom Headers",
"description": "Custom HTTP headers to include with requests.",
"required": false,
"component": "Nested",
"width": "full",
"order": 1,
"properties": {
"Authorization": {
"type": "string",
"title": "Authorization",
"description": "Authorization header for authenticated requests.",
"default": "",
"component": "Input",
"placeholder": "Bearer token or Basic auth",
"width": "full",
"order": 1
},
"Accept": {
"type": "string",
"title": "Accept",
"description": "Accept header to specify preferred content types.",
"default": "*/*",
"component": "Select",
"enum": [
{
"label": "All Types",
"value": "*/*",
"description": "Accept any content type",
"default": true
},
{
"label": "HTML",
"value": "text/html",
"description": "HTML documents only"
},
{
"label": "JSON",
"value": "application/json",
"description": "JSON data only"
},
{
"label": "XML",
"value": "application/xml,text/xml",
"description": "XML documents"
},
{
"label": "Plain Text",
"value": "text/plain",
"description": "Plain text content"
}
],
"width": "half",
"order": 2
},
"Accept-Language": {
"type": "string",
"title": "Accept Language",
"description": "Preferred language for content.",
"default": "en-US,en;q=0.9",
"component": "Input",
"placeholder": "en-US,en;q=0.9,zh-CN;q=0.8",
"width": "half",
"order": 3
},
"Referer": {
"type": "string",
"title": "Referer",
"description": "Referer header for the request.",
"default": "",
"component": "Input",
"placeholder": "https://example.com",
"width": "full",
"order": 4
},
"Cookie": {
"type": "string",
"title": "Cookie",
"description": "Cookie header for session management.",
"default": "",
"component": "Input",
"placeholder": "session=abc123; preference=light",
"width": "full",
"order": 5
}
}
},
"user_agent": {
"type": "string",
"title": "User Agent",
"description": "User agent string to identify the fetcher.",
"default": "GraphRAG-Fetcher/1.0",
"component": "Select",
"enum": [
{
"label": "Default GraphRAG",
"value": "GraphRAG-Fetcher/1.0",
"description": "Default GraphRAG fetcher user agent",
"default": true
},
{
"label": "Chrome Desktop",
"value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
"description": "Chrome browser on Windows"
},
{
"label": "Firefox Desktop",
"value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:120.0) Gecko/20100101 Firefox/120.0",
"description": "Firefox browser on Windows"
},
{
"label": "Safari macOS",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Version/17.0 Safari/537.36",
"description": "Safari browser on macOS"
},
{
"label": "Mobile Chrome",
"value": "Mozilla/5.0 (Linux; Android 10; SM-G973F) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Mobile Safari/537.36",
"description": "Chrome browser on Android"
},
{
"label": "Custom",
"value": "",
"description": "Enter custom user agent below"
}
],
"width": "full",
"order": 2
},
"custom_user_agent": {
"type": "string",
"title": "Custom User Agent",
"description": "Custom user agent string (used when 'Custom' is selected above).",
"default": "",
"component": "Input",
"placeholder": "Your custom user agent string",
"width": "full",
"order": 3
},
"timeout": {
"type": "integer",
"title": "Timeout (seconds)",
"description": "Request timeout in seconds.",
"default": 300,
"minimum": 5,
"maximum": 3600,
"component": "InputNumber",
"width": "half",
"order": 4
},
"follow_redirects": {
"type": "boolean",
"title": "Follow Redirects",
"description": "Automatically follow HTTP redirects.",
"default": true,
"component": "Switch",
"width": "half",
"order": 5
},
"max_redirects": {
"type": "integer",
"title": "Max Redirects",
"description": "Maximum number of redirects to follow.",
"default": 10,
"minimum": 0,
"maximum": 50,
"component": "InputNumber",
"width": "half",
"order": 6
},
"verify_ssl": {
"type": "boolean",
"title": "Verify SSL",
"description": "Verify SSL certificates for HTTPS requests.",
"default": true,
"component": "Switch",
"width": "half",
"order": 7
}
}
}

188
yao/data/kb/providers/fetcher/http/zh-cn.json Normal file
View file

@ -0,0 +1,188 @@
{
"id": "__yao.http",
"title": "HTTP 网络获取器",
"description": "使用 HTTP/HTTPS 协议从网络 URL 获取内容。支持自定义头部、用户代理配置和超时设置，用于可靠的网络内容检索。",
"required": [],
"properties": {
"headers": {
"type": "object",
"title": "自定义头部",
"description": "请求中包含的自定义 HTTP 头部。",
"required": false,
"component": "Nested",
"width": "full",
"order": 1,
"properties": {
"Authorization": {
"type": "string",
"title": "授权",
"description": "用于身份验证请求的授权头部。",
"default": "",
"component": "Input",
"placeholder": "Bearer token 或 Basic auth",
"width": "full",
"order": 1
},
"Accept": {
"type": "string",
"title": "接受",
"description": "指定首选内容类型的接受头部。",
"default": "*/*",
"component": "Select",
"enum": [
{
"label": "所有类型",
"value": "*/*",
"description": "接受任何内容类型",
"default": true
},
{
"label": "HTML",
"value": "text/html",
"description": "仅 HTML 文档"
},
{
"label": "JSON",
"value": "application/json",
"description": "仅 JSON 数据"
},
{
"label": "XML",
"value": "application/xml,text/xml",
"description": "XML 文档"
},
{
"label": "纯文本",
"value": "text/plain",
"description": "纯文本内容"
}
],
"width": "half",
"order": 2
},
"Accept-Language": {
"type": "string",
"title": "接受语言",
"description": "内容的首选语言。",
"default": "zh-CN,zh;q=0.9,en;q=0.8",
"component": "Input",
"placeholder": "zh-CN,zh;q=0.9,en-US;q=0.8",
"width": "half",
"order": 3
},
"Referer": {
"type": "string",
"title": "引用页",
"description": "请求的引用页头部。",
"default": "",
"component": "Input",
"placeholder": "https://example.com",
"width": "full",
"order": 4
},
"Cookie": {
"type": "string",
"title": "Cookie",
"description": "用于会话管理的 Cookie 头部。",
"default": "",
"component": "Input",
"placeholder": "session=abc123; preference=light",
"width": "full",
"order": 5
}
}
},
"user_agent": {
"type": "string",
"title": "用户代理",
"description": "用于标识获取器的用户代理字符串。",
"default": "GraphRAG-Fetcher/1.0",
"component": "Select",
"enum": [
{
"label": "默认 GraphRAG",
"value": "GraphRAG-Fetcher/1.0",
"description": "默认 GraphRAG 获取器用户代理",
"default": true
},
{
"label": "Chrome 桌面版",
"value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
"description": "Windows 上的 Chrome 浏览器"
},
{
"label": "Firefox 桌面版",
"value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:120.0) Gecko/20100101 Firefox/120.0",
"description": "Windows 上的 Firefox 浏览器"
},
{
"label": "Safari macOS",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Version/17.0 Safari/537.36",
"description": "macOS 上的 Safari 浏览器"
},
{
"label": "移动端 Chrome",
"value": "Mozilla/5.0 (Linux; Android 10; SM-G973F) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Mobile Safari/537.36",
"description": "Android 上的 Chrome 浏览器"
},
{
"label": "自定义",
"value": "",
"description": "在下方输入自定义用户代理"
}
],
"width": "full",
"order": 2
},
"custom_user_agent": {
"type": "string",
"title": "自定义用户代理",
"description": "自定义用户代理字符串（在上方选择\"自定义\"时使用）。",
"default": "",
"component": "Input",
"placeholder": "您的自定义用户代理字符串",
"width": "full",
"order": 3
},
"timeout": {
"type": "integer",
"title": "超时时间（秒）",
"description": "请求超时时间，以秒为单位。",
"default": 300,
"minimum": 5,
"maximum": 3600,
"component": "InputNumber",
"width": "half",
"order": 4
},
"follow_redirects": {
"type": "boolean",
"title": "跟随重定向",
"description": "自动跟随 HTTP 重定向。",
"default": true,
"component": "Switch",
"width": "half",
"order": 5
},
"max_redirects": {
"type": "integer",
"title": "最大重定向次数",
"description": "要跟随的最大重定向次数。",
"default": 10,
"minimum": 0,
"maximum": 50,
"component": "InputNumber",
"width": "half",
"order": 6
},
"verify_ssl": {
"type": "boolean",
"title": "验证 SSL",
"description": "验证 HTTPS 请求的 SSL 证书。",
"default": true,
"component": "Switch",
"width": "half",
"order": 7
}
}
}

221
yao/data/kb/providers/fetcher/mcp/en.json Normal file
View file

@ -0,0 +1,221 @@
{
"id": "__yao.mcp",
"title": "MCP Tool Fetcher",
"description": "Model Context Protocol (MCP) fetcher that uses external tools for content retrieval. Allows integration with custom fetching tools through the MCP interface.",
"required": ["id", "tool"],
"properties": {
"id": {
"type": "string",
"title": "MCP Server ID",
"description": "Identifier of the MCP server to use for fetching content.",
"default": "",
"component": "Input",
"width": "half",
"order": 1
},
"tool": {
"type": "string",
"title": "Tool Name",
"description": "Name of the MCP tool to invoke for content fetching.",
"default": "fetch",
"component": "Input",
"width": "half",
"order": 2
},
"arguments_mapping": {
"type": "object",
"title": "Arguments Mapping",
"description": "Mapping of fetch parameters to MCP tool arguments.",
"required": false,
"component": "Nested",
"width": "full",
"order": 3,
"properties": {
"url": {
"type": "string",
"title": "URL Argument",
"description": "MCP tool argument name for the URL to fetch.",
"default": "url",
"component": "Input",
"width": "half",
"order": 1
},
"method": {
"type": "string",
"title": "Method Argument",
"description": "MCP tool argument name for HTTP method.",
"default": "method",
"component": "Input",
"width": "half",
"order": 2
},
"headers": {
"type": "string",
"title": "Headers Argument",
"description": "MCP tool argument name for HTTP headers.",
"default": "headers",
"component": "Input",
"width": "half",
"order": 3
},
"timeout": {
"type": "string",
"title": "Timeout Argument",
"description": "MCP tool argument name for request timeout.",
"default": "timeout",
"component": "Input",
"width": "half",
"order": 4
},
"user_agent": {
"type": "string",
"title": "User Agent Argument",
"description": "MCP tool argument name for user agent string.",
"default": "user_agent",
"component": "Input",
"width": "half",
"order": 5
},
"follow_redirects": {
"type": "string",
"title": "Follow Redirects Argument",
"description": "MCP tool argument name for redirect following option.",
"default": "follow_redirects",
"component": "Input",
"width": "half",
"order": 6
}
}
},
"result_mapping": {
"type": "object",
"title": "Result Mapping",
"description": "Mapping of MCP tool response fields to fetch results.",
"required": false,
"component": "Nested",
"width": "full",
"order": 4,
"properties": {
"content": {
"type": "string",
"title": "Content Field",
"description": "Response field containing the fetched content.",
"default": "content",
"component": "Input",
"width": "half",
"order": 1
},
"url": {
"type": "string",
"title": "URL Field",
"description": "Response field containing the final URL (after redirects).",
"default": "url",
"component": "Input",
"width": "half",
"order": 2
},
"status_code": {
"type": "string",
"title": "Status Code Field",
"description": "Response field containing the HTTP status code.",
"default": "status_code",
"component": "Input",
"width": "half",
"order": 3
},
"headers": {
"type": "string",
"title": "Headers Field",
"description": "Response field containing the response headers.",
"default": "headers",
"component": "Input",
"width": "half",
"order": 4
},
"content_type": {
"type": "string",
"title": "Content Type Field",
"description": "Response field containing the content type.",
"default": "content_type",
"component": "Input",
"width": "half",
"order": 5
},
"content_length": {
"type": "string",
"title": "Content Length Field",
"description": "Response field containing the content length.",
"default": "content_length",
"component": "Input",
"width": "half",
"order": 6
},
"metadata": {
"type": "string",
"title": "Metadata Field",
"description": "Response field containing additional metadata.",
"default": "metadata",
"component": "Input",
"width": "half",
"order": 7
},
"error": {
"type": "string",
"title": "Error Field",
"description": "Response field containing error information.",
"default": "error",
"component": "Input",
"width": "half",
"order": 8
}
}
},
"notification_mapping": {
"type": "object",
"title": "Notification Mapping",
"description": "Mapping for MCP notification handling during fetching.",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"progress": {
"type": "string",
"title": "Progress Notification",
"description": "Notification type for fetch progress updates.",
"default": "progress",
"component": "Input",
"width": "half",
"order": 1
},
"status": {
"type": "string",
"title": "Status Notification",
"description": "Notification type for status changes.",
"default": "status",
"component": "Input",
"width": "half",
"order": 2
},
"redirect": {
"type": "string",
"title": "Redirect Notification",
"description": "Notification type for redirect events.",
"default": "redirect",
"component": "Input",
"width": "half",
"order": 3
},
"error": {
"type": "string",
"title": "Error Notification",
"description": "Notification type for error events.",
"default": "error",
"component": "Input",
"width": "half",
"order": 4
}
}
}
}
}

221
yao/data/kb/providers/fetcher/mcp/zh-cn.json Normal file
View file

@ -0,0 +1,221 @@
{
"id": "__yao.mcp",
"title": "MCP 工具获取器",
"description": "模型上下文协议（MCP）获取器，使用外部工具进行内容检索。允许通过 MCP 接口与自定义获取工具集成。",
"required": ["id", "tool"],
"properties": {
"id": {
"type": "string",
"title": "MCP 服务器 ID",
"description": "用于获取内容的 MCP 服务器标识符。",
"default": "",
"component": "Input",
"width": "half",
"order": 1
},
"tool": {
"type": "string",
"title": "工具名称",
"description": "用于内容获取的 MCP 工具名称。",
"default": "fetch",
"component": "Input",
"width": "half",
"order": 2
},
"arguments_mapping": {
"type": "object",
"title": "参数映射",
"description": "获取参数到 MCP 工具参数的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 3,
"properties": {
"url": {
"type": "string",
"title": "URL 参数",
"description": "要获取的 URL 的 MCP 工具参数名称。",
"default": "url",
"component": "Input",
"width": "half",
"order": 1
},
"method": {
"type": "string",
"title": "方法参数",
"description": "HTTP 方法的 MCP 工具参数名称。",
"default": "method",
"component": "Input",
"width": "half",
"order": 2
},
"headers": {
"type": "string",
"title": "头部参数",
"description": "HTTP 头部的 MCP 工具参数名称。",
"default": "headers",
"component": "Input",
"width": "half",
"order": 3
},
"timeout": {
"type": "string",
"title": "超时参数",
"description": "请求超时的 MCP 工具参数名称。",
"default": "timeout",
"component": "Input",
"width": "half",
"order": 4
},
"user_agent": {
"type": "string",
"title": "用户代理参数",
"description": "用户代理字符串的 MCP 工具参数名称。",
"default": "user_agent",
"component": "Input",
"width": "half",
"order": 5
},
"follow_redirects": {
"type": "string",
"title": "跟随重定向参数",
"description": "重定向跟随选项的 MCP 工具参数名称。",
"default": "follow_redirects",
"component": "Input",
"width": "half",
"order": 6
}
}
},
"result_mapping": {
"type": "object",
"title": "结果映射",
"description": "MCP 工具响应字段到获取结果的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 4,
"properties": {
"content": {
"type": "string",
"title": "内容字段",
"description": "包含获取内容的响应字段。",
"default": "content",
"component": "Input",
"width": "half",
"order": 1
},
"url": {
"type": "string",
"title": "URL 字段",
"description": "包含最终 URL（重定向后）的响应字段。",
"default": "url",
"component": "Input",
"width": "half",
"order": 2
},
"status_code": {
"type": "string",
"title": "状态码字段",
"description": "包含 HTTP 状态码的响应字段。",
"default": "status_code",
"component": "Input",
"width": "half",
"order": 3
},
"headers": {
"type": "string",
"title": "头部字段",
"description": "包含响应头部的响应字段。",
"default": "headers",
"component": "Input",
"width": "half",
"order": 4
},
"content_type": {
"type": "string",
"title": "内容类型字段",
"description": "包含内容类型的响应字段。",
"default": "content_type",
"component": "Input",
"width": "half",
"order": 5
},
"content_length": {
"type": "string",
"title": "内容长度字段",
"description": "包含内容长度的响应字段。",
"default": "content_length",
"component": "Input",
"width": "half",
"order": 6
},
"metadata": {
"type": "string",
"title": "元数据字段",
"description": "包含附加元数据的响应字段。",
"default": "metadata",
"component": "Input",
"width": "half",
"order": 7
},
"error": {
"type": "string",
"title": "错误字段",
"description": "包含错误信息的响应字段。",
"default": "error",
"component": "Input",
"width": "half",
"order": 8
}
}
},
"notification_mapping": {
"type": "object",
"title": "通知映射",
"description": "获取期间 MCP 通知处理的映射。",
"required": false,
"component": "Nested",
"width": "full",
"order": 5,
"properties": {
"progress": {
"type": "string",
"title": "进度通知",
"description": "获取进度更新的通知类型。",
"default": "progress",
"component": "Input",
"width": "half",
"order": 1
},
"status": {
"type": "string",
"title": "状态通知",
"description": "状态变更的通知类型。",
"default": "status",
"component": "Input",
"width": "half",
"order": 2
},
"redirect": {
"type": "string",
"title": "重定向通知",
"description": "重定向事件的通知类型。",
"default": "redirect",
"component": "Input",
"width": "half",
"order": 3
},
"error": {
"type": "string",
"title": "错误通知",
"description": "错误事件的通知类型。",
"default": "error",
"component": "Input",
"width": "half",
"order": 4
}
}
}
}
}

1
yao/data/kb/providers/fetcher/zh-cn.json
View file

@ -1 +0,0 @@
{}