Build a reversible token map for anonymization
Create a structured token substitution table and sanitized copy of sensitive documents to enable clean round-trip anonymization.
You getToken Substitution Table, anonymized Text and reversion Verification
Prompt
Act as an information security analyst. Create a systematic token replacement map for the text below so it can be safely processed and later restored.
- Identify all direct and indirect identifiers: human names, organizations, locations, dollar figures, and unique software codenames.
- Assign consistent, deterministic surrogate tokens (such as ORG_1, PERSON_A, PRODUCT_X).
- Ensure every occurrence of an entity uses the exact same token throughout.
- Never reveal the original sensitive entity inside the sanitized text block.
Output format:
1. Token Substitution Table (markdown table with columns: Surrogate Token, Entity Type, Original Value)
2. Anonymized Text (complete text rewritten using only the surrogate tokens)
3. Reversion Verification (one-sentence confirmation that all tokens match one-to-one)
Original text:Example output
The model receives the original text, so paste only material your data policy allows that tool to see. Keep the substitution table on your own machine and use it to restore names in the final reply yourself. Works in ChatGPT, Claude and Gemini.
Make it repeatable
Keep using this prompt without pasting it in every time.
Prompt details
- By
- Prompts-GPT editors
- Published
Did this prompt work?
Keep exploring
More like this
Audit text for indirect personal identifiers
Scan drafted text for subtle quasi-identifiers, niche titles, and contextual details that could de-anonymize individuals.
Evaluate task safety and privacy for AI drafting
Assess whether a specific task or text draft is safe to process with an AI model based on privacy, accuracy, and professional stakes.
Turn messy notes into structured comparison tables
Convert unorganized text or raw notes into clean, copy-ready tables with standardized headers and no conversational fluff.
Compress bloated instructions while preserving accuracy
Trim wordy system prompts down to their essential constraints. Use this to reduce token consumption, lower API latency, and eliminate confusing prompt contradictions.
Scrub sensitive names and secrets from text
Remove personal names, client details, passwords, and API keys from raw logs or text while keeping technical context intact.
Spot formatting regressions across model updates
Audit outputs from upgraded AI models against your strict formatting rules. Use it to catch broken JSON, lost headers, or leaking chain-of-thought tokens.
- Create a reusable library of personal AI prompts
- Evaluate two responses against a structured rubric
- Match target output format using provided examples
- Process pasted text between clear start and end markers
- Audit drafts for confidently phrased factual hallucinations
- Build a regression test suite for workflow instructions