Suidasu
AI-OCR for invoices, faxes, handwriting and photos. Extract structured data with custom layouts.
Should I use this
Quality & Safety
Findings (1)
- LOWin extract_data
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"suidasu": {
"url": "https://api.suidasu.com/mcp"
}
}
}Remote endpoints
https://api.suidasu.com/mcpstreamable-httpWhat it can do
Tool inventory
Tools (21)
🟢extract_data(extraction_mode, image_url, project_id)
Extract structured data from an image using a predefined layout. Uses AI cascade strategy for optimal cost-accuracy balance.
Input Schema
{
"type": "object",
"properties": {
"extraction_mode": {
"description": "Extraction mode: 'ocr' for text recognition (default), 'image_recognition' for visual analysis of images without text",
"enum": [
"ocr",
"image_recognition"
],
"type": "string"
},
"image_url": {
"description": "URL of the image to process (HTTPS URL)",
"type": "string"
},
"project_id": {
"description": "UUID of the project (required)",
"type": "string"
}
},
"required": [
"project_id",
"image_url"
]
}🟢list_layouts(limit, offset)
List all available extraction layouts with their field definitions
Input Schema
{
"type": "object",
"properties": {
"limit": {
"description": "Maximum number of layouts to return (default: 20)",
"type": "integer"
},
"offset": {
"description": "Number of layouts to skip (for pagination)",
"type": "integer"
}
}
}🟢get_extraction(extraction_id)
Get the result of a previous extraction by ID. Returns extracted data, a signed URL for downloading the original image (valid for 1 hour), and error_message if the extraction failed (e.g. inappropriate_content, prohibited_content, low_confidence).
Input Schema
{
"type": "object",
"properties": {
"extraction_id": {
"description": "UUID of the extraction to retrieve",
"type": "string"
}
},
"required": [
"extraction_id"
]
}🟢list_extractions(date_from, date_to, latest_only, limit, offset, ...)
List OCR extractions filtered by project and date range. Failed extractions include error_message field indicating the failure reason (e.g. scan.error.inappropriate_content, scan.error.low_confidence).
Input Schema
{
"type": "object",
"properties": {
"date_from": {
"description": "Start date in RFC3339 format, e.g. 2024-01-01T00:00:00Z (required)",
"type": "string"
},
"date_to": {
"description": "End date in RFC3339 format, e.g. 2024-01-31T23:59:59Z (required)",
"type": "string"
},
"latest_only": {
"description": "If true, exclude extractions that were superseded by a later re-extraction of the same file (superseded_by is set). Default: false",
"type": "boolean"
},
"limit": {
"description": "Maximum number of results to return (default: 20)",
"type": "integer"
},
"offset": {
"description": "Number of results to skip for pagination (default: 0)",
"type": "integer"
},
"project_id": {
"description": "Project UUID (required)",
"type": "string"
}
},
"required": [
"project_id",
"date_from",
"date_to"
]
}🟢list_projects(limit, offset)
List all projects with their layouts and field definitions
Input Schema
{
"type": "object",
"properties": {
"limit": {
"description": "Maximum number of projects to return (default: 20)",
"type": "integer"
},
"offset": {
"description": "Number of projects to skip (for pagination)",
"type": "integer"
}
}
}🔴update_extraction(extracted_data, extraction_id)
Update the extracted data of an extraction result
Input Schema
{
"type": "object",
"properties": {
"extracted_data": {
"description": "New extracted data to replace the existing data",
"type": "object"
},
"extraction_id": {
"description": "UUID of the extraction to update",
"type": "string"
}
},
"required": [
"extraction_id",
"extracted_data"
]
}🔴delete_extraction(extraction_id)
Delete an extraction result. A parent (root) ID deletes the whole upload (parent and all records); a record ID deletes only that record and renumbers the rest. Deleting an extraction that superseded an older extraction of the same file makes the older one current again.
Input Schema
{
"type": "object",
"properties": {
"extraction_id": {
"description": "UUID of the extraction to delete",
"type": "string"
}
},
"required": [
"extraction_id"
]
}🟡create_project(default_extraction_mode, description, icon_color, layout_id, layout_source, ...)
Create a new project with layout options
Input Schema
{
"type": "object",
"properties": {
"default_extraction_mode": {
"description": "Default extraction mode for this project. Default: 'ocr'",
"enum": [
"ocr",
"image_recognition"
],
"type": "string"
},
"description": {
"description": "Description of the project (optional)",
"type": "string"
},
"icon_color": {
"description": "Icon color (e.g., 'blue', 'red', 'green'). Default: 'blue'",
"type": "string"
},
"layout_id": {
"description": "UUID of an existing layout to use for this project (required). Create a layout first using create_layout, import_template, or create_layout_from_recipe.",
"type": "string"
},
"layout_source": {
"description": "How to use the layout: 'shared' (reference existing layout, default) or 'copy' (create a copy of the layout)",
"enum": [
"shared",
"copy"
],
"type": "string"
},
"name": {
"description": "Name of the project",
"type": "string"
}
},
"required": [
"name",
"layout_id"
]
}🔴update_project(default_extraction_mode, description, icon_color, is_active, name, ...)
Update an existing project
Input Schema
{
"type": "object",
"properties": {
"default_extraction_mode": {
"description": "New default extraction mode",
"enum": [
"ocr",
"image_recognition"
],
"type": "string"
},
"description": {
"description": "New description of the project",
"type": "string"
},
"icon_color": {
"description": "New icon color",
"type": "string"
},
"is_active": {
"description": "Whether the project is active",
"type": "boolean"
},
"name": {
"description": "New name of the project",
"type": "string"
},
"project_id": {
"description": "UUID of the project to update",
"type": "string"
}
},
"required": [
"project_id"
]
}🔴delete_project(project_id)
Delete a project (soft delete)
Input Schema
{
"type": "object",
"properties": {
"project_id": {
"description": "UUID of the project to delete",
"type": "string"
}
},
"required": [
"project_id"
]
}🟡create_layout(description, fields, name)
Create a new layout with field definitions. The created layout can be assigned to a project via create_project.
Input Schema
{
"type": "object",
"properties": {
"description": {
"description": "Description of the layout (optional)",
"type": "string"
},
"fields": {
"description": "Array of field definitions",
"items": {
"properties": {
"auto_field": {
"description": "Auto-populated field type: extraction_id, image_url, shot_datetime, shot_latitude, shot_longitude",
"enum": [
"",
"extraction_id",
"image_url",
"shot_datetime",
"shot_latitude",
"shot_longitude"
],
"type": "string"
},
"default_value": {
"description": "Default/fixed value for the field (used when is_fixed_value is true)",
"type": "string"
},
"display_name": {
"description": "Display name for the field",
"type": "string"
},
"field_type": {
"description": "Type of the field. Default: 'text'",
"enum": [
"text",
"number",
"date",
"time",
"bool",
"email",
"phone",
"currency"
],
"type": "string"
},
"is_fixed_value": {
"description": "If true, this field has a fixed value and AI will not extract it",
"type": "boolean"
},
"name": {
"description": "Field name (used as key in extracted data)",
"type": "string"
}
},
"required": [
"name"
],
"type": "object"
},
"type": "array"
},
"name": {
"description": "Name of the layout (required)",
"type": "string"
}
},
"required": [
"name",
"fields"
]
}🔴update_layout(fields, layout_id)
Update an existing layout (replaces all fields)
Input Schema
{
"type": "object",
"properties": {
"fields": {
"description": "New array of field definitions (replaces existing fields; must not be empty)",
"items": {
"properties": {
"auto_field": {
"description": "Auto-populated field type: extraction_id, image_url, shot_datetime, shot_latitude, shot_longitude",
"enum": [
"",
"extraction_id",
"image_url",
"shot_datetime",
"shot_latitude",
"shot_longitude"
],
"type": "string"
},
"default_value": {
"description": "Default/fixed value for the field (used when is_fixed_value is true)",
"type": "string"
},
"display_name": {
"description": "Display name for the field",
"type": "string"
},
"field_type": {
"description": "Type of the field. Default: 'text'",
"enum": [
"text",
"number",
"date",
"time",
"bool",
"email",
"phone",
"currency"
],
"type": "string"
},
"is_fixed_value": {
"description": "If true, this field has a fixed value and AI will not extract it",
"type": "boolean"
},
"name": {
"description": "Field name (used as key in extracted data)",
"type": "string"
}
},
"required": [
"name"
],
"type": "object"
},
"type": "array"
},
"layout_id": {
"description": "UUID of the layout to update",
"type": "string"
}
},
"required": [
"layout_id",
"fields"
]
}🔴delete_layout(layout_id)
Delete a layout (soft delete)
Input Schema
{
"type": "object",
"properties": {
"layout_id": {
"description": "UUID of the layout to delete",
"type": "string"
}
},
"required": [
"layout_id"
]
}🟢generate_recipe(comment, field_language, image_url)
Analyze a sample file and generate layout proposals. Accepts images, PDF, CSV, Excel (xlsx), Markdown, and plain text (e.g. a layout specification document). AI automatically detects the document type and suggests appropriate field definitions. Requires Standard plan or higher. Daily limit: 50 uses.
Input Schema
{
"type": "object",
"properties": {
"comment": {
"description": "Optional user comment to guide AI analysis (e.g., 'I want to extract invoice line items', 'Focus on product names and prices'). Max 500 characters.",
"type": "string"
},
"field_language": {
"description": "Language for generated field descriptions and layout names. Supported: ja (Japanese, default), en (English), es (Spanish), fr (French), zh (Simplified Chinese).",
"enum": [
"ja",
"en",
"es",
"fr",
"zh"
],
"type": "string"
},
"image_url": {
"description": "URL of the file to analyze (HTTPS). Supports image/PDF samples and CSV/Excel/Markdown/plain-text documents. Only single file URLs are supported; folder URLs are not accepted.",
"type": "string"
}
},
"required": [
"image_url"
]
}🟡create_layout_from_recipe(description, fields, name)
Create a new layout from a recipe proposal. Use the output from generate_recipe to create a layout.
Input Schema
{
"type": "object",
"properties": {
"description": {
"description": "Description of the layout (optional)",
"type": "string"
},
"fields": {
"description": "Array of field definitions from recipe proposal",
"items": {
"properties": {
"description": {
"description": "Field description",
"type": "string"
},
"format": {
"description": "Format hint (date, time, email, phone, url) for string type",
"type": "string"
},
"name": {
"description": "Field name (snake_case)",
"type": "string"
},
"type": {
"description": "Field type (string, number, integer, boolean, array)",
"type": "string"
}
},
"required": [
"name"
],
"type": "object"
},
"type": "array"
},
"name": {
"description": "Name of the layout",
"type": "string"
}
},
"required": [
"name",
"fields"
]
}🟢export_extractions(date_from, date_to, format, project_id)
Export extraction data to CSV or Excel format. Returns base64-encoded file data.
Input Schema
{
"type": "object",
"properties": {
"date_from": {
"description": "Start date filter in RFC3339 or YYYY-MM-DD format (optional)",
"type": "string"
},
"date_to": {
"description": "End date filter in RFC3339 or YYYY-MM-DD format (optional)",
"type": "string"
},
"format": {
"description": "Export format: 'csv', 'excel', or 'jsonl'. Default: 'csv'",
"enum": [
"csv",
"excel",
"jsonl"
],
"type": "string"
},
"project_id": {
"description": "UUID of the project to export extractions from (required)",
"type": "string"
}
},
"required": [
"project_id"
]
}🟢list_templates
List all available system templates (business_card, invoice, receipt, etc.) with their field definitions
Input Schema
{
"type": "object",
"properties": {}
}⚪import_template(name, template_id)
Import a system template as a new layout for your account
Input Schema
{
"type": "object",
"properties": {
"name": {
"description": "Custom name for the imported layout (optional, defaults to template name)",
"type": "string"
},
"template_id": {
"description": "ID of the system template to import (required)",
"enum": [
"business_card",
"invoice",
"receipt",
"drivers_license",
"purchase_order",
"delivery_note",
"inspection_report",
"daily_work_report"
],
"type": "string"
}
},
"required": [
"template_id"
]
}🟢get_usage_stats
Get current month's usage statistics including extraction count, remaining free extractions, and success rate
Input Schema
{
"type": "object",
"properties": {}
}🟢get_billing_info(year_month)
Get billing information for a specific month including invoice amount and billable extraction count from Stripe
Input Schema
{
"type": "object",
"properties": {
"year_month": {
"description": "Target month in YYYY-MM format (e.g., 2026-03)",
"type": "string"
}
},
"required": [
"year_month"
]
}🟢check_fields(document_type, fields, language)
Validate field definitions for AI extractability. Returns confidence score and improvement suggestions.
Input Schema
{
"type": "object",
"properties": {
"document_type": {
"description": "Type of document these fields are for (e.g., 'invoice', 'receipt')",
"type": "string"
},
"fields": {
"description": "Array of field definitions to validate",
"items": {
"properties": {
"display_name": {
"description": "Human-readable display name",
"type": "string"
},
"field_type": {
"description": "Type of the field. Default: 'text'",
"enum": [
"text",
"number",
"date",
"time",
"bool",
"email",
"phone",
"currency"
],
"type": "string"
},
"name": {
"description": "Field name (snake_case recommended)",
"type": "string"
}
},
"required": [
"name"
],
"type": "object"
},
"type": "array"
},
"language": {
"description": "Language for suggestions. Default: 'ja'",
"enum": [
"ja",
"en"
],
"type": "string"
}
},
"required": [
"fields"
]
}Community
Evidence