DocPro · Agent
PTAS DocPro KYC & Identity Agent
Identity fields read, checked against what the applicant submitted, and redacted on export so onboarding moves without leaking data.
What the agent does
Read, Checked, Then Masked
Identity documents are the one category where extracting less, and holding it for less time, is the better outcome.
Read
Document classification
The type and issuing authority are identified first, so the right fields are looked for and the right checks applied.
Read
Field extraction
Name, document number, dates of birth and issue, expiry and address, taken from the layout each issuer uses.
Read
Machine-readable zone
Where a document carries an MRZ, it is read and reconciled against the printed fields as an independent check.
Check
Applicant cross-check
Extracted values are compared with what the applicant typed, and differences are reported rather than overwritten.
Check
Expiry and validity
Documents expired at submission, or expiring inside your review window, are flagged before they reach a reviewer.
Protect
Redaction on export
Sensitive values are masked when data leaves the platform, at the field level you specify rather than all or nothing.
Cross-checking
What the Form Says Against What the Document Says
Most onboarding problems are not forgeries. They are transposed digits, a married name, an address that moved. Each one needs surfacing, not silently correcting.
Held internally against exported · Illustrative example
Masking is set per field and per destination, so a KYC reviewer and a billing system receive different versions of the same record.
Field-level redaction
You choose which values leave unmasked.
Expiry flags
Caught at submission, not at audit.
Mismatch detection
Typed details against what the document says.
Privacy controls
Holding Less of It
Data minimisation
Only the fields your process needs are extracted and retained. What is not required is not held in the first place.
Redaction on export
Masking is applied per field on the way out, so downstream systems receive what they need and nothing more.
Access control
Who can see an unredacted identity document is a permission, separate from who can process the application.
Retention limits
Documents are held for the period your policy specifies and then removed, with the deletion recorded.
FAQ
Questions Teams Ask
Where are identity documents stored, and for how long?
In the region you choose, for the period your retention policy specifies, after which they are removed and the deletion is recorded. On-premise and air-gapped deployments are available where a regulator requires the documents never to leave your own infrastructure.
What gets redacted, and can we control it?
Redaction is per field and configured by you. Most teams mask the full document number and date of birth on export while keeping the name and verification result, but the split is yours to set, including different rules for different downstream systems.
Does this decide whether to accept an applicant?
No. It extracts the fields, cross-checks them against what was submitted, and reports expiry and tamper signals. Whether that adds up to an acceptable applicant is a decision your policy makes, with a person on it where your policy says so.
Which identity documents are supported?
The common government-issued types (passports, national identity cards and driving licences) across the issuers you onboard from, plus any additional document your process requires, defined during onboarding like any other custom document.
Keep reading
Related Capabilities
Vendor Management
The same verification discipline applied to suppliers.
Read more →Banking & Finance
KYC at the volumes financial services onboard at.
Read more →Intelligent Document Processing
The classification and confidence layer underneath.
Read more →DocPro
Every other document type the same engine reads.
Read more →Next step
Bring Your Retention Policy to the Call
For identity work, where the data lives and how long you keep it matters more than extraction accuracy.