social-media-manager
AdvisoryAudited by Static analysis on May 7, 2026.
Overview
No suspicious patterns detected.
Findings (0)
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.
If connected to real social media accounts, the agent could help draft or trigger public replies, likes, follows, or direct messages that affect a brand or personal account.
The skill describes public-facing social engagement actions. This is aligned with the social media manager purpose, but such actions should be explicitly approved if the agent is later given tools that can post, message, like, or follow accounts.
type EngagementAction = | { type: 'reply'; template: string } | { type: 'like' } | { type: 'follow-back'; criteria: string[] } | { type: 'dm'; template: string }Require user approval before any live posting, messaging, liking, following, or account interaction, and restrict connected tools to the intended accounts.
A bad post, incorrect claim, or off-brand message could be scheduled or adapted for several platforms, increasing reputational impact.
The skill supports multi-platform content generation and scheduling. This is purpose-aligned, but mistakes in generated or scheduled content could be amplified across multiple public channels if published without review.
platforms: Platform[]; // Target platforms ... schedule?: Date; // Optional scheduling
Review generated calendars and platform-specific posts before publishing, especially for multi-platform campaigns or scheduled content.