Turn noisy trend inputs into ranked, publishable decisions.
Priority order:
1) demand signal quality
2) audience fit
3) monetization fit
4) execution speed
Gather 10–30 candidate signals from:
Record provenance for each signal:
source_type (official/community/internal)source_link (if available)captured_atconfidence (high/medium/low)If live endpoints are unavailable, run fallback mode using recent internal patterns and clearly label output as mode: fallback.
For each topic, standardize:
topicplatform_fit (TikTok / YouTube / Instagram)intent_type (learn / compare / buy / troubleshoot / inspiration)freshness (hot / warm / evergreen)audience_fit (1–5)monetization_fit (1–5)difficulty (1–5)Merge near-duplicate topics before scoring.
Use:
priority_score = (audience_fit 0.35) + (freshness_score 0.25) + (monetization_fit 0.25) + (execution_speed 0.15)
Mapping:
freshness_score: hot=5, warm=3, evergreen=2execution_speed = 6 - difficultyReturn:
Include data_confidence for each topic (high/medium/low).
Copyright (c) 2026 Razestar.
This skill is provided under CC BY-NC-SA 4.0 for non-commercial use.
You may reuse and adapt it with attribution to Razestar, and share derivatives
under the same license.
Commercial use requires a separate paid commercial license from Razestar.
No trademark rights are granted.
共 1 个版本