Security audit

Social Content Converter

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly generates social-media copy, but its instructions also describe automated posting through logged-in accounts without clear confirmation controls.

Install only if you want both content generation and possible social-platform publishing workflows. Treat generation and publishing as separate steps, review every generated post manually, and do not let an agent use logged-in Douyin, Xiaohongshu, or Bilibili sessions unless you explicitly intend to publish.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger phrases are broad everyday requests like '改写内容' and '生成各平台文案', which can cause the skill to activate unintentionally during normal conversation. In context, accidental activation is more dangerous because the skill also advertises file generation and optional publishing workflows, increasing the chance of unintended content processing or downstream external actions.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill describes automated publishing workflows to Douyin, Xiaohongshu, and Bilibili, including using logged-in browser sessions, but does not clearly warn users that it may perform external account actions. This is dangerous because accidental or misunderstood use could post content under the user's authenticated accounts, causing reputational harm, unwanted disclosures, or policy violations.

Static analysis

No suspicious patterns detected.