Social Media content writing
PassAudited by ClawScan on May 1, 2026.
Overview
This is an instruction-only social media content and operations guide, with no code or credential access; the main things to notice are its public-influence and audience-profile guidance.
This skill appears safe to install as an instruction-only content strategy guide. Before using it for real campaigns, keep human approval over publishing, verify factual claims, respect platform rules and privacy expectations, and avoid covert manipulation or undisclosed promotional content.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Content created with this skill may influence public perception or community sentiment.
The skill explicitly covers public-opinion guidance and topic marketing. This is aligned with social media operations, but it can affect how audiences perceive events or brands.
适合事件报道、舆论引导及话题营销 ... 在负面评论出现时快速响应，引导舆论走向
Use it for truthful, transparent communications; avoid deceptive engagement tactics, undisclosed sponsorships, or manipulative crisis messaging.
A factual mistake, inappropriate message, or rights issue could spread across multiple channels if reused without review.
The skill encourages repurposing one source across many formats and platforms. That is purpose-aligned, but an error or misleading claim in the source material could be amplified broadly.
一次采集，多种生成：将同一核心素材转化为图文、短视频、直播、音频等多种形态
Fact-check and approve core source material before adapting it for multiple platforms.
Audience analytics or saved content could include sensitive, stale, or inaccurate information if retained and reused carelessly.
The skill contemplates reusable audience profiles and content/material libraries. No storage mechanism is provided, but if a user implements this workflow, retained profile or source data should be scoped and verified.
画像构建：基于数据勾勒用户画像（年龄、地域、兴趣） ... 建立“热点日历”与“素材库”
Use aggregated or consented analytics where possible, limit retention, and treat comments, trend data, and stored materials as untrusted until verified.