Anonymize your data. Safely, in your browser
Hash, mask, redact or pseudonymize sensitive columns in CSV, Excel and TSV files. The motor runs locally; nothing is uploaded. Optional reversible mapping lets you restore originals later.
Drop a file to anonymize
CSV, TSV, TXT or Excel. Runs in your browser, never uploaded.
Supported: .csv, .tsv, .txt, .xlsx, .xls
Why anonymize before sharing data
Whenever you share a dataset with an AI assistant, a teammate, a contractor or an external tool, the PII in that dataset is at risk. Anonymization replaces names, emails, phone numbers, DNIs, IBANs and other identifiers with synthetic or hashed values. So you can use the data without exposing the people behind it.
- Paste anonymized data into ChatGPT, Claude or any LLM without leaking customer info
- Share a dataset with developers or analysts who don't need the real names
- Meet GDPR and AI Act pseudonymization expectations for internal analytics
- Reverse the anonymization later with the mapping file. It never leaves your machine
How it works
- 1Drop your fileCSV, TSV or Excel. Parsing happens in your browser; the data is never uploaded.
- 2Pick columns and methodsDetected PII columns are pre-suggested. Choose hash, redact, fake values, or reversible pseudonymization per column.
- 3Run free preview, then fullPreview the first 100 rows free to verify the result. Run the full anonymization (free up to 2,000 rows).
- 4Download file + mappingGet the anonymized file plus a mapping dictionary (JSON or CSV). Store the mapping if you want to reverse later.
Four methods, one tool
Hash is one-way and deterministic (the same input produces the same SHA-256 hash). Redact replaces values with a placeholder like [REDACTED]. Faker generates realistic synthetic values per type (email, name, phone, Spanish DNI, IBAN, city). Pseudonymize emits incremental tokens (EMAIL_0001, EMAIL_0002…) per column.
Faker (reversible) and Pseudonymize produce a mapping file you can use to restore originals. Hash and Redact are deliberately one-way. Pick the right method for your use case.
Need to restore the original values later?
Use the de-anonymize tool: upload the anonymized file together with the mapping dictionary and the original values come back. Works for pasted text from an LLM too.
Open the de-anonymize tool →Related tools
Frequently asked questions
Does my data leave the browser?+
No. The anonymization engine runs entirely in a Web Worker inside your browser. We only persist the metadata needed for the paid tier gate (row counts, file sizes). Never the file contents.
Which methods are reversible?+
Pseudonymize is always reversible. Faker is reversible when you set 'reversible: true' on the column. Hash and Redact are deliberately one-way. Once applied, the original value cannot be recovered.
How do I restore the original data?+
Open the de-anonymize tool and upload the anonymized file together with the mapping file. The tool replaces every mapped token with the original value and gives you a restored file.
Is this GDPR compliant?+
Pseudonymization satisfies GDPR Article 4(5). Strong anonymization with one-way methods (hash, redact) goes further and exits GDPR scope when irreversible. We recommend reading the GDPR pseudonymization page for specifics.
What's the price?+
Free up to 2,000 rows. Above that, the same tier table as Compare and Merge: $3 up to 25k, $7 up to 100k, $15 up to 500k, $29 up to 1M rows. One-time payment, no subscription.
Can I anonymize SQL dump files?+
Not yet. The current engine supports CSV, TSV, TXT and Excel. SQL dump anonymization is on the roadmap but needs careful parsing to preserve INSERT structure.