Security audit

Blog 一文多发

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed content-rewriting helper for turning FridayParts blog summaries into social posts, with no executable code or hidden system access.

Safe to install for drafting social copy from FridayParts blog summaries. Review generated posts before publishing, and be aware that broad triggers like blog rewriting or multi-platform content may activate the skill in adjacent writing tasks.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger phrases are generic and broad, which can cause the skill to activate in conversations that only loosely mention blog rewriting or multi-platform posting. Unintended invocation can lead to prompt hijacking of workflow, inappropriate content transformation, or accidental processing of user content in the wrong context, especially in systems where skills auto-route based on trigger matching.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger phrases are generic content-repurposing terms such as '一文多发', 'blog改写', and '多平台内容', which can match many ordinary user requests outside the intended FridayParts blog workflow. This can cause unintended activation and prompt hijacking of unrelated tasks, leading the agent to apply brand-specific instructions or rigid output formatting when the user did not intend to invoke this skill.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.