Peer Review Pass

The problem

Drafting research-heavy content often leads to internal blind spots where you stop seeing your own logical leaps or misattributed details. Standard AI reviews tend to either agree with you too much or flatten your authorial voice with unnecessary hedging.

What it does

• Executes a six-point structured audit focusing on false attribution, unattested details, and source conflation.
• Identifies interpretive overclaiming only when genuine expert disagreement exists, preserving your strong authorial stance.
• Uncovers two types of omissions: evidence that contradicts your claim and the "missing" best evidence that would strengthen it.
• Triggers an isolated adversarial read designed to catch errors that survived the initial drafting context.
• Generates a non-destructive report that flags issues by location without rewriting your prose.

Why this beats prompting it yourself

Generic prompts for fact-checking usually result in vague feedback or "hallucinated" style critiques. This skill uses a two-stage architecture that forces the LLM to separate structured logical checks from a context-free adversarial pass, mimicking a professional editorial pipeline.

Use cases

• Audit technical whitepapers for comparative accuracy and source integrity.
• Review historical or research-based articles for subtle attribution errors.
• Fact-check argumentative essays to ensure they wouldn't crumble under expert scrutiny.
• Verify that complex case studies don't conflate distinct events or data points.

Known limitations

This is not a systematic primary-source database verification or a style audit. It focuses on the logic, attribution, and structural integrity of claims rather than prose aesthetics.