One API call removes personal data from text, JSON, and CSV before it reaches LLMs, vendors, or test environments. Nothing is stored. Nothing leaves your infrastructure.
curl -X POST https://api.comply-tech.co.uk/api/v1/anonymise \
-H "X-Api-Key: demo-key-complytech" \
-H "Content-Type: application/json" \
-d '{
"content": "Contact Sarah Mitchell at sarah.m@gmail.com or 07700 900123",
"contentType": "text",
"strategy": "Redact",
"frameworks": ["GDPR"]
}'Paste this into your terminal. Takes 30 seconds. No signup required.
Every support ticket, chat message, and customer record you send to OpenAI, Anthropic, or Google contains personal data you probably shouldn't be sharing. The LLM doesn't need it. Your compliance team definitely doesn't want you sharing it.
Hi, my name is Sarah Mitchell and I placed an order last week but haven't received it. My order number is #48291 and my email is sarah.mitchell@gmail.com. I live at 14 Beechwood Avenue, Manchester, M20 3FJ. Can someone help?
Hi, my name is [NAME REDACTED] and I placed an order last week but haven't received it. My order number is #48291 and my email is [EMAIL REDACTED]. I live at [ADDRESS REDACTED]. Can someone help?
The LLM still produces the same quality summary and suggested reply. It cares about the problem described, not who described it.
# Step 1: Strip PII
clean = requests.post("https://api.comply-tech.co.uk/api/v1/anonymise",
headers={"X-Api-Key": "your-key", "Content-Type": "application/json"},
json={"content": ticket, "contentType": "text",
"strategy": "Redact", "frameworks": ["GDPR"]}
).json()["anonymisedContent"]
# Step 2: Send clean data to your LLM
reply = openai.chat.completions.create(
model="gpt-4", messages=[{"role": "user", "content": clean}]
)Someone in your team is about to email a spreadsheet full of customer names and emails to an analytics vendor. It happens every week. The API strips PII and keeps the business data intact.
Name,Email,TotalSpend,LastOrder James O'Brien,james@gmail.com,£2340.50,2024-11-15 Priya Sharma,priya@outlook.com,£890.00,2024-12-01 Michael Chen,m.chen@yahoo.co.uk,£5100.75,2024-10-28
Name,Email,TotalSpend,LastOrder [NAME REDACTED],[EMAIL REDACTED],£2340.50,2024-11-15 [NAME REDACTED],[EMAIL REDACTED],£890.00,2024-12-01 [NAME REDACTED],[EMAIL REDACTED],£5100.75,2024-10-28
Replace with [REDACTED] labels. Best when PII is never needed again.
Partial visibility: j***@gmail.com. Useful for recognition.
Deterministic fake data. Same input always produces same output.
Your staging database is full of real customer data because synthetic data is too clean. Pseudonymise it instead: real structure, fake identities, relationships intact.
id,name,email,notes 1,James O'Brien,james@gmail.com,"Late delivery to 42 Victoria Rd" 2,Priya Sharma,priya@outlook.com,"VIP customer, Birmingham office" 3,Michael Chen,m.chen@yahoo.co.uk,"Invoice to michael.chen@work.com"
id,name,email,notes 1,David Thompson,sarah.j@outlook.com,"Late delivery to [ADDRESS REDACTED]" 2,Emily Watson,mark.w@gmail.com,"VIP customer, Birmingham office" 3,Sarah Williams,j.taylor@yahoo.co.uk,"Invoice to [EMAIL REDACTED]"
You're sending PII to a third party to detect PII. For many organisations, that defeats the purpose. ComplyTech processes with zero data retention: nothing is stored, nothing is logged.
You'll spend weeks setting up ML models, building infrastructure, and maintaining it. ComplyTech is one API call. No spaCy, no model hosting, no Python environment on a finance laptop.
You'll catch emails and phone numbers but miss names, addresses, and PII embedded in free-text fields. And you'll maintain it forever. One customer named "Will May" and your name detector breaks.
Processed in memory, never stored. Nothing is persisted after the API response.
Fast enough to sit in front of your LLM call without adding noticeable delay.
GDPR, HIPAA, CCPA, PCI-DSS, and 6 more. Every response includes a compliance report.
Send any format. Get the same structure back with PII stripped from every field.
Hit the button to call the live API with real data.
Click below to call the live API
Testing the API with your own data? 2,000 fields/month, no card required.
Processing 5,000 support tickets a month through your LLM? That's Starter.
Refreshing a 50,000-row staging database weekly? That's Pro.
Processing millions of fields across all jurisdictions with an SLA? Let's talk.
For side projects & prototypes
No credit card required
For growing products
Billed monthly, excl. tax
For scaling companies
Billed monthly, excl. tax
For large-scale deployments
Custom pricing based on fields/month
Prices are in GBP and exclusive of applicable taxes. Higher tiers unlock additional frameworks and higher rate limits.
A "field" is one cell in a CSV, one value in JSON, or 1KB of plain text (minimum 1 field).
No. LLMs care about the problem described, not who described it. A support ticket with [NAME REDACTED] produces the same quality summary and suggested reply as one with the real name.
No. Data is processed in memory and returned. Nothing is persisted, logged, or retained after the API response. Zero data retention by design.
Billed by fields processed. A field is one cell in a CSV, one value in JSON, or 1KB of plain text (minimum 1 field). A 1,000-row CSV with 7 columns uses 7,000 fields. The free tier includes 2,000 fields/month.
Ten frameworks: GDPR, HIPAA, CCPA, PCI-DSS, LGPD, PIPEDA, DPDPA, POPIA, PDPA, and APPI. Higher tiers unlock additional frameworks. You can combine multiple in a single request.
Text, JSON, and CSV. Send a support ticket, a JSON API payload, or a full CSV export. The structure is preserved; only PII values are replaced.
We monitor regulatory changes and update processing rules automatically. All customers benefit immediately, with no code changes on your end.
Grab your API key below, or paste the demo curl command into your terminal right now.
Or try the demo key first, no signup needed.
curl -X POST https://api.comply-tech.co.uk/api/v1/anonymise \
-H "X-Api-Key: demo-key-complytech" \
-H "Content-Type: application/json" \
-d '{"content":"Sarah Mitchell, sarah@gmail.com, 07700 900123","contentType":"text","strategy":"Redact","frameworks":["GDPR"]}'