Security audit

代码

Security checks across malware telemetry and agentic risk

Overview

This appears to be an IRIS/ObjectScript formatter, but it requests broad file, command, and API authority without clear limits.

Review this skill before installing. Use it only in a limited workspace, avoid giving it secrets or API keys, and review any file changes or shell commands before allowing them; its formatting guidance may be useful, but its declared authorities and API/command claims are broader than the stated purpose.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Description-Behavior Mismatch

High

Confidence: 95% confidence
Finding: The skill is presented as an IRIS/ObjectScript formatter and reviewer, but the documentation expands its scope to generic file handling, API integration, and command execution. This scope creep increases the chance that an agent will invoke the skill in broader contexts and grant it capabilities unrelated to its stated purpose, violating least privilege and enabling misuse.

Context-Inappropriate Capability

High

Confidence: 98% confidence
Finding: The manifest exposes generic exec alongside read/write/glob/grep for a skill whose stated purpose is code formatting and review. Unrestricted command execution gives the skill the ability to run arbitrary shell commands if invoked by an agent, which can lead to filesystem modification, data exfiltration, or execution of untrusted code paths.

Context-Inappropriate Capability

Medium

Confidence: 87% confidence
Finding: The documentation advertises external API integration even though the skill's main purpose is local code formatting/review. That encourages broader trust and may prompt agents or users to supply secrets or permit networked actions that are unnecessary for the declared task, expanding attack surface without clear justification.

Intent-Code Divergence

Medium

Confidence: 93% confidence
Finding: The security section claims commands are restricted to a whitelist, but the declared tool is generic exec with no visible technical enforcement. This mismatch is dangerous because reviewers or downstream agents may rely on a control that does not actually exist, resulting in overtrust and unsafe execution behavior.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The invocation guidance is broad enough to match general code generation, development, debugging, testing, and deployment requests rather than only IRIS formatting/review. Overbroad matching can cause the skill to be selected in contexts where its powerful tools are unnecessary, increasing the chance of misuse or unintended command/file operations.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.