Frontier-grade Claude models with agentic workflows, strong coding, and enterprise guardrails delivered via developer console, web app, and cloud partners.
anonymise.online
€4.99/moStarting price
0Popularity
About anonymise.online
anonymise.online provides a reversible data-hiding layer for LLM workflows, designed to protect sensitive text before it is pasted into AI systems. The tool detects personal data, credentials, business identifiers and confidential text, replacing them with readable placeholders such as PERSON, EMAIL or ADDRESS. Users can review the anonymized output and export a private restore map to reverse the process later. The detection process occurs locally using predefined rules and offline models, ensuring no external AI providers receive or inspect the text. The workflow involves three steps: protecting prompts before sending them to AI, using the anonymized version in AI tools, and restoring original values afterward using the private map. It supports both text and PDF files, with usage limits varying by plan. The service emphasizes data control and privacy, positioning itself as a lightweight protection layer rather than a VPN or AI gateway.
Key features
- Text and PDF anonymization
- Local detection using rules and offline models
- Readable placeholder labels (PERSON, EMAIL, ADDRESS)
- Manual phrase selection and custom placeholder creation
- Private restore map export
- No external AI provider access to text
- Prompt protection before AI use
- Post-AI restoration of original values
Use cases
- Masking sensitive data in AI prompts before sharing
- Protecting confidential documents before processing with LLMs
- Reversibly anonymizing business or personal identifiers in text
Pros
- Local detection with no external AI inspection
- Reversible anonymization with private restore maps
- Supports both text and PDF files
- No permanent account required for access
- Manual phrase control for custom placeholders
Cons
- Limited free tier (500 characters, 2 anonymizations/day)
- No permanent account option for premium plans
- Usage limits per plan (characters, anonymizations, PDFs)
- Single payment plan expires after one use
Frequently asked questions about anonymise.online
What does anonymise.online do?
Anonymise.online is a reversible data-hiding layer for LLM workflows that detects and masks sensitive text such as personal data, credentials, and business identifiers before pasting prompts into AI systems. It replaces sensitive values with readable placeholders like PERSON or EMAIL and allows users to restore original text later using a private map.
Who should use anonymise.online?
The tool suits individuals and organizations that need to protect sensitive information when using AI tools, including developers, researchers, legal professionals, and businesses handling confidential data. It is designed for anyone who wants to maintain privacy while leveraging AI capabilities.
How does anonymise.online handle my data?
Detection occurs locally using predefined rules and offline models, ensuring no external AI providers receive or inspect the text. The service does not send data to third-party AI models like ChatGPT or Claude for detection, and users retain full control over their data and restore maps.
Does anonymise.online require an account?
No, anonymise.online offers no-signup plans for text and PDF protection. Users can access free or paid features by purchasing access codes sent via email, without creating a permanent account. Premium plans are managed through Stripe subscriptions.
What file formats does anonymise.online support?
The tool supports both text and PDF files for anonymization. Free plans allow limited PDF processing, while paid plans offer higher limits for both text and PDF anonymization.
How do I get started with anonymise.online?
Users can start by visiting the website, entering text or uploading a PDF, and following the three-step workflow: protect the text, use the anonymized version in AI tools, and restore original values using the private map. No permanent account is required for most plans.
