Here's what happens when you drop a document.
From the moment a file enters Archeko to the moment it's filed and queryable — every step is automatic. Your only job is the upload.
CAPTURE
Upload once. Everything else is handled.
Receive
Upload a document from any source — drag and drop, email forward, mobile scan, API. Optionally attach a note for context the AI can use during processing.
Read
Archeko extracts the full text content from the document, whether it's a native PDF, a scanned image, or a photo from a job site. Handwriting, stamps, and low-quality scans are handled.
Classify
The AI identifies the document type — Invoice, Contract, Permit, Receipt, Delivery Note — based on its content, not its filename. Your configured document types define the categories.
Extract
Structured fields are pulled automatically: dates, amounts, vendor names, project references, contract numbers. The fields depend on the document type and your configuration.
Name
A canonical filename is generated according to your naming template. `scan_003_FINAL.pdf` becomes `2024-11-15_BuildCorp_Invoice_INV-2847.pdf`. Consistent, searchable, human-readable.
File
The document is placed in the correct folder path based on your filing rules — by entity, project, document type, and date. It is linked to the appropriate cost object automatically.
QUERY
Talk to your archive.
Archeko exposes your document archive through MCP (Model Context Protocol) — the open standard for connecting AI assistants to external data.
Connect it to Claude or ChatGPT, and ask questions in plain language.
- User: Show me all invoices from BuildCorp in the last 6 months.
- User: What's the total cost on Project Alpine?
- User: Find the signed lease for the Maribor warehouse.
- User: Which permits expire before June?
Every answer references the actual source documents. You are not trusting the AI's memory — you are querying structured, verified data with a full audit trail.
YOUR RULES
Configured to your business. Not a template.
Archeko does not impose a fixed structure. During onboarding, we configure the system to match how your business actually works:
Document types
Invoice, Contract, Permit, Lease, Receipt, or whatever your business uses. You define the categories — the AI learns them.
Fields per type
The data extracted depends on the document type. An invoice extracts amount, vendor, date, and invoice number. A permit extracts expiry date, issuing authority, and project reference. You decide what matters.
Filing path templates
Where documents land in your folder structure. Define paths by entity, project, year, document type — or any combination.
Filename templates
How documents are named. Set the pattern once, and every document follows it automatically.
Cost objects
Projects, departments, entities, or any cost structure you track. Every document is linked to its financial context.
The AI follows your rules. When your business changes, the configuration changes with it. Reconfiguration is included.
THE DIFFERENCE
What changes.
Before Archeko
- ×Documents scattered across Drive, email, and WhatsApp
- ×Manual renaming:
Invoice_FINAL_v2_SIGNED.pdf - ×Filing depends on who remembers where things go
- ×Finding a document means searching, scrolling, asking around
- ×Audit prep takes days of reconstruction
- ×No connection between documents and financial context
With Archeko
- Every document in one structured archive
- Canonical naming:
2024-11-15_BuildCorp_Invoice_INV-2847.pdf - Filing is automatic and rule-based
- Finding a document means asking a question in plain language
- Audit trail is built-in and always current
- Every document linked to its project, entity, and cost object
See it work with your own documents.
The best way to understand Archeko is to try it. Upload a real invoice, contract, or receipt and watch it process in seconds.
Start your free trial