Security audit

Workflow Essentials

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly an automation workflow guide, but it requests broad command and file permissions and describes operational create/export/callback behavior without clear boundaries.

Review this before installing if you expected a documentation-only skill. Do not let it run commands, modify files, post content, create invoices, connect accounts, or enable external automations unless you explicitly requested that action and can inspect the exact target workflow.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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

Findings (6)

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The skill is presented as a workflow methodology guide, but its troubleshooting section instructs the agent to execute a ping command and perform network diagnostics. That expands behavior from passive guidance into active command execution, which can be abused to trigger unintended shell activity or reconnaissance in environments where the skill is trusted to be documentation-only.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The manifest and description frame this skill as methodology and configuration guidance, yet the body claims executable behavior using input_params for create/query/export operations. This mismatch can cause an agent or user to over-trust the skill and allow actions beyond its stated purpose, increasing the chance of unauthorized writes, exports, or workflow manipulation.

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The example section depicts actual shell execution via a bash block and reports successful operational completion, despite the skill being described as a guidance-oriented methodology document. Examples strongly influence agent behavior, so this can normalize command execution in contexts where users expected only instructional output.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The skill claims coverage across a very broad set of keywords and scenarios without narrow trigger conditions or exclusions. Over-broad invocation scope increases the chance that an agent will apply this skill in unrelated contexts, potentially surfacing exec/write-capable behaviors when the user did not request operational actions.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill advertises read/exec/write tools and supports callback URLs, but does not provide clear user-facing warnings about shell execution, file modification, or outbound communication. In a guidance-style skill, these capabilities materially raise risk because users may not realize the skill can write files, execute commands, or interact with external endpoints.

Missing User Warnings

Low

Confidence: 92% confidence
Finding: The troubleshooting guidance instructs command execution for connectivity testing without clear warning that shell commands may be run. Even a simple diagnostic command can become a precedent for broader command execution, especially in agent environments that interpret skill content as operational guidance.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.