Write & communicate

Audit content for repetition, vagueness, and machine-like writing

Diagnose observable writing problems without unreliable claims about whether AI produced the text

8 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

Audit the supplied content for observable repetition, vagueness, and formulaic writing.

Exact draft, purpose, format, status, and known authorship context:
[draft]

Audience, effect, approved samples, terminology, invariant facts, and style boundaries:
[voice]

Diagnosis or revision scope, depth, priorities, length, and protected text:
[scope]

Do not claim to detect whether AI wrote the text; authorship cannot be reliably inferred from style alone. Instead cite exact locations and diagnose observable effects: repeated claims or structure, empty abstraction, unsupported generalization, generic transitions, excessive symmetry, uniform sentence rhythm, redundant conclusion, vague actors or verbs, nominalization, cliché, unexplained jargon, false precision, abrupt tone shifts, and voice mismatch. Distinguish intentional repetition from accidental redundancy and conventional genre language from genuine weakness. For each issue state evidence, reader effect, severity, confidence, and a revision operation. Preserve facts, meaning, quotations, terminology, and author intent. If revision is authorized, provide a minimal revised passage plus a change log; otherwise do not rewrite. End with a priority order and checks requiring human judgment of voice and factual fidelity.
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

Draft and context
LinkedIn draft: In today's fast-paced world, effective handoffs are more important than ever. Handoffs are important because they help teams work effectively. By optimizing handoffs, organizations can unlock efficiency, enhance collaboration, and drive meaningful outcomes. First, clarify ownership. Second, clarify expectations. Third, clarify communication. Ultimately, clear handoffs are the key to success. In conclusion, organizations should prioritize clear handoffs to achieve success.
Audience and voice reference
Operations leads who prefer direct, evidence-aware advice. Approved voice sample starts with a concrete workflow symptom, uses 'owner,' 'accepter,' and 'acceptance evidence,' and avoids broad business claims. Invariant idea: a handoff needs an owner and accepter; no measured outcome evidence supplied.
Review scope
Diagnose and minimally revise. Under 700 characters. Prioritize vagueness and repetition. Preserve the distinction between owner and accepter. Do not add statistics, personal anecdotes, or claims about success.

Example output

Authorship note: these patterns do not establish whether AI or a person wrote the draft.

High priority • Opening: “today's fast-paced world” provides no situation or decision. Reader effect: delays the usable point. Replace with a concrete stalled-review symptom. High confidence. • Outcome claims: “unlock efficiency, enhance collaboration, and drive meaningful outcomes” are abstract and unsupported. Reader effect: sounds promotional but gives nothing to verify. Remove rather than strengthen. • Conclusion: “key to success” and the final sentence repeat the same unsupported claim. Keep one practical CTA instead.

Medium priority • Three-item sequence repeats “clarify” while hiding who decides what. Replace generic nouns with owner, accepter, acceptance evidence, and escalation. The parallel structure itself is not a problem; its categories are vague.

Minimal revision “Work is marked ready, but nobody knows who can accept it. Check one recurring handoff: Who owns the current step? Who accepts it next? What evidence counts as accepted? Who resolves a delay or disagreement? A status field can display these answers, but the team still has to define them. Audit one handoff before adding another reminder.”

Change log: removed two generic openings, three unsupported outcome phrases, and duplicate success conclusions; replaced vague categories with the approved operational terms; preserved the owner-versus-accepter distinction. Human check: confirm “status field” fits the account voice and that the final CTA matches the intended product context.

Why this works

  1. 1

    Observable diagnosis improves prose without pretending style can prove authorship.

  2. 2

    Reader-effect and confidence labels prevent mechanical deletion of intentional choices.

Check the result

  • Does every diagnosis cite exact text and an observable reader effect?

  • Does the audit avoid unsupported AI-authorship judgments?

  • Do revisions preserve facts, voice, intentional repetition, and protected text?

More ways to explore

Where this recipe fits

Keep the work moving