Security audit

编码格式工具(免费版)

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward encoding and decoding reference with command examples; its write-capable examples need care but match the stated purpose.

Install this as a utility skill for explicit encoding, decoding, JWT inspection, or file-format conversion tasks. Before allowing command examples that use output redirection or sed -i, check the target filename and consider using a temporary copy to avoid overwriting useful data.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Behavioral ASTexec() Call, eval() Call, Dynamic Import
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The skill description is unusually broad and generic, advertising applicability to many "encoding formats" tasks without clear trigger constraints or boundaries on when the agent should invoke it. In an agent ecosystem, this can cause over-selection of an exec-capable skill for loosely related prompts, increasing the chance that shell examples or transformations are applied to sensitive user data or files unnecessarily.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The examples include commands that write files and modify files in place, such as redirecting decoded output to image.png and using sed -i to alter an existing file, but they are not accompanied by prominent safety warnings or confirmation requirements. In an exec-enabled skill, an agent may copy these commands into a live workspace, causing unintended overwrites or destructive edits if filenames or inputs are user-controlled or misunderstood.

Static analysis

No suspicious patterns detected.