Available now · Deterministic · Hosted in Germany

Detect European personal data before it leaves your workflow.

GDPR Shield adds a personal-data classification gate to any LLM, AI agent, or text workflow that sends text through its HTTP API. It classifies IBANs, tax IDs, national identifiers, credentials, and review-worthy patterns across 10 European countries. No LLM calls. No input-text retention.

$ curl -X POST shield.promptandbuild.de/v2/classify Start for 29 EUR

IBAN · Tax ID · BSN · Personnummer · NIF · Rijksregister · PESEL · Codice Fiscale

shield $ "Herr Mueller, IBAN DE89370400440532013000" DETECTED IBAN, PERSON_SALUTATION shield $ "Mozartstrasse 5, 1010 Wien" REVIEW SUSPECTED_ADDRESS shield $ "Pan Kowalski, PESEL 44051401458" DETECTED PERSONAL_ID_NUMBER, PERSON_SALUTATION shield $ "Docker nginx traefik deployment log" CLEAR No configured identifier detected shield $ "NIF: 123456789, Rua Augusta 234" DETECTED TAX_ID, SUSPECTED_ADDRESS shield $ "ghp_ABCDEFghijKLMNopqrSTUVwxyz123456" DETECTED CREDENTIALS
28-second product demo

See GDPR Shield classify sensitive text.

Watch the API identify European personal-data patterns, return a clear decision, and keep the original text out of logs and storage.

Start with GDPR Shield English captions are on by default and can be toggled in the video controls.
Measured, not implied

A security claim you can inspect.

The engine is evaluated with versioned regression, adversarial, Unicode-evasion, API, database, startup, concurrency, and webhook-security suites.

0.903precision on the internal red-team benchmark
0.966recall on the internal red-team benchmark
0.933F1 score with documented limitations
544engine regression, penetration, and Unicode-evasion checks
The problem

Personal data ends up where it shouldn't

Text pipelines process customer messages, support tickets, form submissions, and AI agent outputs. Personal identifiers slip through into logs, analytics, training data, and third-party services.

AI agents and LLM pipelines

LLM inputs and outputs regularly contain names, IBANs, and ID numbers. Without a filter, personal data flows into prompts, logs, and downstream systems.

Support and CRM systems

Customers paste account numbers, tax IDs, and personal details into chat and email. Agents forward these into tickets, analytics, and reporting tools.

Data pipelines and ETL

Batch processing moves text between systems. A single unfiltered field can propagate personal data across databases, data warehouses, and backups.

Use cases

Where teams use GDPR Shield

Add a single API call to your pipeline. The response tells you whether the text contains personal data and what kind.

CI/CD guardrail

Scan AI agent outputs before they reach production. Flag text classified as DETECTED for human review. Automate what you can, escalate what you must.

Log sanitization

Check log entries before they hit your logging infrastructure. Classify and tag entries so personal data can be handled according to your retention policies.

Form and chat filtering

Classify incoming text from forms, chat, and email. Route messages containing personal identifiers to appropriate handling workflows.

How it works

Three tiers, clear boundaries

Every text is classified into one of three tiers. Mathematical checksums (Mod97, Luhn, Elfproef) validate identifiers. Humans only review what the system marks as uncertain.

DETECTED (confirmed)

Contains verified personal data. IBAN validated via Mod97, credit card via Luhn, tax ID via country-specific checksum. High confidence.

REVIEW (review)

Potentially contains personal data. Two capitalized words in sequence, address-like patterns without full context. Flagged for human decision.

CLEAR (clear)

No configured identifier was detected. This is not proof that the text contains no personal data; choose a policy that matches your risk.

10
Countries
400
Regression tests
131
Attack tests
13
Unicode evasions
Why GDPR Shield

What makes this different

Built for teams that process European text and need to know what's in it before it moves downstream.

No input-text retention

The managed API runs on German infrastructure. Input text is processed in memory and is not written to application logs or the database. Metadata is pseudonymized for security and billing. An AVV/DPA is available for business customers.

Math, not AI guesswork

IBANs are validated with Mod97. Credit cards with Luhn. Dutch BSN with Elfproef. Polish PESEL with weighted checksums. Pattern matching plus mathematical proof reduces false positives.

Built for European formats

Each country module knows its own ID formats, address patterns, and salutation conventions. Not a generic regex list, but country-specific detection logic covering 10 EU/EEA countries.

Predictable local engine

No LLM inference and no third-party classifier calls. The engine uses deterministic patterns and checksum functions; end-to-end latency still depends on your network and deployment region.

Coverage

10 countries, real validation

Each country module combines format rules, context checks, and the applicable validation algorithms. Not every data category has a mathematical checksum; uncertain matches are returned as REVIEW.

🇩🇪

Germany

Mod97Luhn IBAN, Tax ID, SSN, license plate, ID card

🇫🇷

France

Mod97 social security number, business registration, health insurance card, identity card

🇳🇱

Netherlands

Elfproef BSN, KvK, postal code, driver's license

🇪🇸

Spain

Mod23 DNI/NIE, corporate tax ID, social security number, passport

🇮🇹

Italy

Checksum Codice Fiscale, Partita IVA, ID card

🇵🇱

Poland

Weighted PESEL, NIP, REGON, identity card

🇦🇹

Austria

Mod97 SOCIAL_SECURITY_NUMBER, VAT_ID, tax number

🇧🇪

Belgium

Mod97 National register, enterprise number, eID

🇵🇹

Portugal

Weighted NIF, citizen card, social security number, passport

🇸🇪

Sweden

Luhn Personnummer, org number, passport

Choose the right layer

Managed European detection without building a DLP platform.

Shield is intentionally narrower than a full enterprise DLP suite and easier to adopt than operating and tuning a generic open-source toolkit.

CapabilityGeneric cloud PII APISelf-hosted toolkitGDPR Shield
Ready-to-use managed APIYesCustomer operatedYes
Ten European country modulesVariesRequires configurationIncluded
National checksum validationVariesRecognizer dependentBuilt in
No LLM inferenceService dependentPossibleGuaranteed by design
No input-text retentionPolicy dependentCustomer controlledManaged-service default
Automatic masking or tokenizationOften availableOften availableNot yet; classification only
Integration

One endpoint, full control

Send JSON, get a classification. Pass country codes to scope the detection. Multiple countries in a single request.

POST /v2/classify
# Request curl -X POST https://shield.promptandbuild.de/v2/classify \ -H "Authorization: Bearer YOUR_KEY" \ -H "Content-Type: application/json" \ -d '{"text": "Herr Mueller, IBAN DE89370400440532013000", "countries": ["de"]}' # Response { "level": "DETECTED", "categories": ["IBAN", "PERSON_SALUTATION"], "countries": ["de"], "ms": 0.3 }
GET /v2/health
GET /v2/stats (Auth)
Pricing

Start classifying in 60 seconds

Shield Starter is available now: subscribe, receive your API key by email, and start classifying. The Agent Safety Bundle is available through coordinated early access.

Starter

29EUR/month
  • 1,000 classifications / day
  • All 10 countries
  • Deterministic engine without LLM calls
  • No input-text retention
  • AVV/DPA available for business use
  • Email support
  • Cancel anytime
Get started

Secure payment via Stripe. Invoice delivered automatically.

Agent Safety Bundle

49EUR/developer/month
  • Everything in Shield Starter, including 1,000 classifications per API key per day
  • GuardRail Pro with its tested Claude Code adapter
  • Advanced action guards derived from real incidents
  • Audit Trail and penetration-test framework
  • Action security plus data security
  • Coordinated onboarding
Request bundle access

Early-access bundle. We confirm fit and onboarding before activation.

Questions before you integrate

Know exactly what Shield does.

Does Shield make an application GDPR compliant?

No. Shield is a technical classifier that supports data-handling controls. Compliance depends on your purpose, legal basis, processes, contracts, retention, access controls, and the rest of your system.

Does CLEAR prove that a text contains no personal data?

No. CLEAR means no configured identifier or review pattern was detected. Apply a policy that matches the risk of your workflow and retain human review where the consequences are high.

Do you store the text we submit?

The managed API processes input text in memory and does not write it to application logs or the database. Pseudonymized metadata is retained only where needed for security and billing.

Does Shield redact or anonymize findings?

Not currently. The API classifies the text and returns categories so your workflow can block, route, tag, or transform it. Automated masking is a planned extension.

Can existing v1 integrations continue to work?

Yes. API v2 is the recommended English contract. API v1 remains available for backward compatibility while existing clients migrate.