Manufacturing RFQ Intake & BOM Preflight

A focused Apify Actor for manufacturers, job shops, contract manufacturers, PCB/PCBA teams, injection molders, sheet-metal shops, and automation workflows that receive RFQs by email.

It converts customer RFQ text plus optional public PDF/text attachments into a structured quotation-intake brief and flags missing or conflicting information before an estimator or engineer spends time reviewing it. An optional BOM CSV preflight catches blank/invalid quantities and duplicate part numbers.

Why this exists

Manufacturing RFQs often arrive as a mix of email text, PDFs, drawings, BOMs, and notes. Sales and estimating teams repeatedly search for the same fields: quantity, material, process, tolerance, finish, lead time, and part/drawing number. Missing details create back-and-forth before quoting.

This Actor is intentionally a preflight assistant, not an engineering authority. It does not approve drawings, select materials, validate GD&T, calculate machinability, or issue a final quotation.

Input

• rfqText: customer email or RFQ notes.
• documentUrls: optional public HTTP(S) PDF/text URLs, up to 10.
• bomCsvUrl: optional public CSV/TSV BOM.
• requiredFields: defaults to quantity, material, process.
• recommendedFields: defaults to tolerance, finish, leadTime.
• maxPagesPerPdf: defaults to 20, maximum 50.
• includeRawText: false by default to minimize retained customer content.

Example:

{
  "rfqText": "Please quote 250 pcs, aluminum 6061-T6, CNC milling, tolerance ±0.05 mm, black anodized, lead time 3 weeks.",
  "requiredFields": ["quantity", "material", "process"]
}

Output

One dataset record per completed preflight, including:

• readiness status and score;
• extracted candidate values;
• missing required fields;
• warnings for multiple candidate values;
• BOM anomaly summary;
• source-processing errors;
• explicit limitations for human review.

Example result:

{
  "status": "ready",
  "readinessScore": 100,
  "missingRequiredFields": [],
  "fields": {
    "quantity": {"primary": "250", "candidates": ["250"]},
    "material": {"primary": "aluminum 6061-T6", "candidates": ["aluminum 6061-T6"]},
    "process": {"primary": "CNC machining", "candidates": ["CNC machining"]}
  }
}

BOM checks

For CSV/TSV BOMs the Actor checks whether recognizable part-number and quantity columns exist, and flags blank part numbers, invalid/non-positive quantities, and duplicate part numbers.

Safety and limits

Only public HTTP(S) document URLs are accepted. URL credentials, non-standard ports, and non-public resolved addresses are rejected, including after redirects. Each document is capped at 8 MB and combined downloads at 20 MB. PDF extraction is text-based; image-only/scanned PDFs may need OCR in a separate workflow.

Final quotation, drawing interpretation, tolerance acceptance, material/process selection, and engineering approval must remain human decisions.

Automation use

The output is suitable for n8n, Make, Zapier, CRM/ERP intake, Google Sheets, or a custom quoting pipeline. A common flow is:

customer email / form → attachment storage → this Actor → estimator review → CRM/ERP record → quote workflow.

Pricing

Release target: $0.15 per completed RFQ preflight using Apify pay-per-event billing. One preflight can include RFQ text, bounded document extraction, and one BOM check. The event is charged only when a result is written.

Development

pip install -r requirements.txt
python -m unittest discover -s tests -v
python -m src