docdigitizer-ai
v0.2.0
Published
Vercel AI SDK tools for the DocDigitizer document processing API
Maintainers
Readme
docdigitizer-ai
Vercel AI SDK tools for the DocDigitizer document processing API.
Installation
npm install docdigitizer-ai aiUsage
import { generateText } from 'ai';
import { openai } from '@ai-sdk/openai';
import { processDocumentTool, healthCheckTool } from 'docdigitizer-ai';
const { text } = await generateText({
model: openai('gpt-4o'),
prompt: 'Process the invoice at /tmp/invoice.pdf and summarize it',
tools: {
processDocument: processDocumentTool({ apiKey: 'dd_live_...' }),
healthCheck: healthCheckTool({ apiKey: 'dd_live_...' }),
},
});Tools
processDocumentTool
Processes a PDF document and returns extracted structured data.
Parameters (defined via Zod schema):
filePath(required): Absolute path to the PDF filepipeline(optional): Pipeline name for processing
Returns:
{
"state": "COMPLETED",
"traceId": "ABC1234",
"numPages": 2,
"extractions": [
{
"documentType": "Invoice",
"confidence": 0.95,
"countryCode": "PT",
"pages": [1, 2],
"data": { "invoice_number": "INV-001", "total": "1000.00" }
}
]
}healthCheckTool
Checks if the DocDigitizer API is available.
Returns:
{ "status": "ok", "message": "I am alive" }Configuration
processDocumentTool({
apiKey: 'dd_live_...', // or set DOCDIGITIZER_API_KEY env var
baseUrl: 'https://custom.api', // optional
timeout: 300000, // optional (ms)
maxRetries: 3, // optional
});License
MIT
