Change interpersonal stance and formality while preserving claims, obligations, uncertainty, and boundaries
4 min setupTested with:ChatGPTReviewed: 2026-08-28
1
Add your context
Your text stays in this browser. AILesson Prompts does not send it to a model or server.
2
Your prompt
Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt
Adjust the tone of the source without changing its facts or practical meaning.
Source:
[text]
Desired warmth, directness, formality, confidence, culture, channel, and relationship:
[tone]
Facts, dates, numbers, responsibility, requests, refusals, uncertainty, quotations, and fixed wording:
[invariants]
Change diction, sentence rhythm, greeting, ordering, and politeness only as needed for the target tone. Preserve who did or must do what, scope, conditions, deadlines, certainty, causality, and consequences. Do not soften a refusal into possibility, intensify a recommendation into an obligation, add praise or apology unsupported by context, or use stereotypes as cultural adaptation. Return: adjusted version; optional slightly warmer and slightly more direct variants; invariant comparison; material rhetorical changes; and source ambiguities that tone editing cannot solve.
Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.
From input to outcome
A worked example
See how concrete context turns this recipe into a usable result
Actual input
Source text
Your request to add four reports to the September release cannot be accepted. The scope was frozen on 12 August. Two reports may be considered for October after the data definitions are approved. Send the definitions to Ana by 5 September if you want them reviewed. The remaining two reports are not planned.
Target tone and relationship
Collaborative but firm email from a product manager to a long-term internal stakeholder. Professional, moderately warm, direct. Do not sound punitive or invite renegotiation of September.
Meaning invariants
Keep: all four reports excluded from September; scope freeze 12 August; only two may be considered for October and only after data-definition approval; definitions go to Ana by 5 September; remaining two are not planned. No apology required and no October commitment.
Example output
Adjusted version: Thanks for setting out the four reporting requests. Because the September scope was frozen on 12 August, we cannot add any of the four reports to that release.
Two of the reports can be considered for October once their data definitions are approved. If you would like those two reviewed, please send the definitions to Ana by 5 September. This is a review step rather than a commitment to the October release. The other two reports are not currently planned.
Invariant audit: all four remain outside September; freeze date, two-report limit, approval condition, Ana, and 5 September are unchanged; October remains possible rather than promised; two reports remain unplanned. Rhetorical changes: acknowledged receipt, led with the fixed reason, clarified that review is not commitment, and replaced “cannot be accepted” with direct release-specific wording. Slightly warmer alternative may add “I’m happy to help clarify the required definitions” only if that help is genuinely available.
Why this works
1
Meaning invariants separate relational style from the commitments that wording carries.
2
Nearby tone variants show the available range without silently moving the message's boundary.
Check the result
Are facts, actors, requests, refusals, dates, conditions, and certainty unchanged?
Does the tone fit the stated relationship and channel without stereotyping?
Could any politeness change be misread as a new commitment or weakened boundary?
ChatGPT