Data Used in Prompts Audit: Is Anything Sensitive Being Sent?

Check what information flows into prompts and models.

Added October 15, 2026. Written with AI assistance and checked for clarity and placeholders; not reviewed by a lawyer, accountant, clinician or other professional. How these prompts are made

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it (what [VERIFY] means).

Audit the data in our prompts. Examples of prompts with placeholders replacing real values (describe what the real values are): [PROMPT_EXAMPLES]. The tools or models used: [TOOLS_USED]. Our data rules: [DATA_RULES].

Classify each data element as public, internal, confidential or restricted, and check whether it is allowed to be sent to the tool according to the rules. Make a table with Prompt, Data element, Class, Allowed, Risk, and Fix (remove, anonymise, summarise, use an approved tool). Count prompts with restricted data. Provide an anonymisation guide with examples, a checklist before pasting, and a short training outline.

Do not paste real sensitive data. Mark contractual and legal points as [VERIFY] with the responsible officer.

Add a worked classification table of five prompts (a marketing brief: internal; a customer email thread with names: confidential; an employee review: restricted; public web copy: public; a contract draft: confidential) with allowed or not allowed for the tool, and a safe version of the employee-review prompt using role titles and no names.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.