Document Ingestion

Upload and process documents for semantic search and RAG.

Supported Formats

Format	Extensions	Notes
PDF	.pdf	Text and scanned (via OCR)
Word	.docx, .doc	Full formatting preserved
Text	.txt	Plain text
Markdown	.md	Rendered to text
HTML	.html	Tags stripped

Upload Documents

Single Document

curl -X POST https://api.gateflow.ai/v1/data/documents \
  -H "Authorization: Bearer gw_prod_..." \
  -F "file=@contract.pdf" \
  -F "classification=confidential" \
  -F "metadata={\"department\": \"legal\", \"year\": 2024}"

Response:

{
  "document_id": "doc_abc123",
  "filename": "contract.pdf",
  "status": "processing",
  "classification": "confidential",
  "created_at": "2024-01-15T10:30:00Z"
}

Batch Upload

curl -X POST https://api.gateflow.ai/v1/data/documents/batch \
  -H "Authorization: Bearer gw_prod_..." \
  -F "files=@doc1.pdf" \
  -F "files=@doc2.pdf" \
  -F "files=@doc3.pdf" \
  -F "classification=internal"

Processing Pipeline

Processing Status

curl https://api.gateflow.ai/v1/data/documents/doc_abc123 \
  -H "Authorization: Bearer gw_prod_..."

Response:

{
  "document_id": "doc_abc123",
  "status": "ready",
  "pages": 15,
  "chunks": 45,
  "pii_detected": true,
  "pii_entities": 3
}

Status Values

Status	Description
uploading	File being received
processing	Being parsed and embedded
ready	Available for search
failed	Processing error

Configuration

Chunking

{
  "chunking": {
    "strategy": "semantic",
    "max_chunk_size": 1000,
    "overlap": 100
  }
}

PII Handling

{
  "pii": {
    "detect": true,
    "action": "redact",
    "types": ["PERSON", "SSN", "PHONE"]
  }
}

Next Steps

• PII Detection - Configure PII handling
• Semantic Search - Query documents