oh-story-long-scan

Collect verifiable ranking samples from your target platform, recording platform, list type, links, access dates, and sample sizes. Then code each sample for genre combination, emotional promise, protagonist starting point, and first payoff to identify recurring reader expectations.

Treat rankings as biased market samples rather than direct answers. Distinguish traffic, paid, new-book, completed, and editorial lists, report sample frequencies first, and flag low-sample combinations as signals rather than confirmed trends.

No. It does not run scraping scripts, browsers, or bypass login restrictions. It only uses sources accessible in the current session, Deep Research results, or user-provided data, and provides a research framework with a checklist when live data is unavailable.

Each candidate should include target readers, core emotion, sample evidence, author strengths, differentiation variables, a minimal opening validation, three-stage upgrade space, and key risks. High-frequency samples are downgraded when author fit or long-term space is weak.

A single sampling only forms hypotheses, not conclusions. Confidence increases through cross-date and cross-list sampling or by deep-reading representative works, and recommendation slots, author fanbases, and trending events must be accounted for as biases.