What this workflow is for
Use this when you’re reviewing a draft before it goes to a client, an editor, or your audience. You’re looking for writing problems you can fix, not a verdict about who wrote it.
A flagged phrase is a place to look. Keep it if it does useful work; change it if it’s padding.
The 5-step workflow at a glance
Collect enough text
Use a full paragraph or draft so rhythm and structure have context.
Name patterns first
Mark the passages that feel padded, repetitive, or over-explained.
Run transparent tooling
Check the actual findings and their explanations against your draft.
Context-check hard
Genre, edits, and language background can explain pattern overlap.
Choose the next action
Revise, investigate, or move on. Never skip straight to accusation.
Step 1: Collect enough text
WROITER accepts 15 words or more. A full paragraph or draft gives you more context for repeated phrasing, rhythm, and structure than a single sentence does.
If you only have a short email, caption, or product blurb, review individual findings in context. A short sample cannot establish authorship.
Step 2: Name the passages that need work
Write down the exact passages that felt suspicious. Can you label them as stock phrases, metronomic rhythm, throat-clearing intro, over-signposting, or pivot crutches?
If your note is only "sounds AI," your evidence is not ready yet. For pattern definitions, use the Pattern Library. For visual recognition, see How to Spot AI Writing.
Step 3: Run a transparent diagnostic
Use a checker that shows the passage and explains the pattern. A percentage gives you no editing instruction.
The WROITER Diagnostic returns named findings and revision instructions. Paste your draft, choose Find patterns & fixes, and compare the findings with your own notes.
Find the AI slop patterns in your text. For a phrase such as “it’s important to note,” the useful question is whether the sentence gets stronger when you start with the claim itself.
Step 4: Context-check before interpretation
Human writing can contain the same patterns. Ask these before you change a passage:
- Genre: academic, legal, policy, and product copy are formal by design.
- Edit depth: heavy editing smooths human variation.
- Language background: second-language writing may prefer safer phrasing.
- Provenance: drafts, revision history, and process notes often matter most.
The Limitations page lists the highest-risk false-positive contexts.
Step 5: Choose the next action
Revise
The finding names a real problem. Use its instruction to edit that passage.
Investigate
Signals are dense and context does not explain them. Ask for process evidence.
Move on
Evidence is weak or context explains the overlap.
When revision instructions are available, choose Copy full prompt. Paste the prompt and the same draft into ChatGPT, Claude, or your usual AI. Ask it to apply the specific fixes, preserve your facts and meaning, and leave unrelated passages alone. Review the result before using it.
If you want WROITER inside the chat, follow the integration walkthrough. If nothing is flagged, there may be no revision prompt; keep editing for accuracy, voice, and purpose.
Common workflow breaks
If this list feels familiar, read the False Positive Hall of Fame before setting policy.
Before you build a policy on this
This workflow is designed to prevent overreaction. But the workflow itself is only as safe as the interpretation rules behind it. Read How the profile is built and how to read it so you know what to make of each pattern family, and review the reliability brief and false positive cases before using any detector output in high-stakes decisions.