Security audit

报告写作

Security checks for vulnerabilities and agentic risk

Overview

This skill is a simple report-writing Markdown skill with disclosed local report output and no artifact-backed hidden, destructive, or exfiltration behavior.

Install only if you want an agent to draft Markdown reports or similar content and potentially create files locally. Ask the agent to confirm the target filename before writing, and avoid granting command execution unless a specific report workflow truly needs it.

SkillSpector

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

High

Confidence: 81% confidence
Finding: The invocation guidance is overly broad and includes unrelated use cases such as marketing copy and content generation, which can cause the agent to apply this skill outside its stated purpose. In a system where the skill has read/exec/write tools, ambiguous triggering increases the chance of unintended file operations or command execution in contexts the user did not clearly request.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The documentation references writing Markdown drafts to a local reports directory but does not present a prominent, explicit warning about local filesystem modification. When a skill advertises write and exec tools, insufficient disclosure can lead users or calling agents to trigger file writes without informed consent, increasing the risk of unintended data creation, overwrite, or workspace pollution.

Static analysis

No suspicious patterns detected.