Security audit

Work Productivity Agent Browser Workflow Helper 002325

Security checks across malware telemetry and agentic risk

Overview

This is a workflow/documentation helper with broad activation wording, but it contains no executable code, hidden persistence, credential handling, or destructive behavior.

Install this only if you want a general Agent Browser workflow helper. Be aware that its broad triggers may cause it to activate for ordinary browser or automation questions, so review whether implicit invocation is appropriate in your environment.

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 (8)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger sentences are broad and partially templated, matching generic requests like needing a practical workflow or help with browser/headless automation. This increases the chance the skill is invoked unintentionally in unrelated contexts, which can cause inappropriate guidance, unsafe automation suggestions, or unexpected browser-workflow actions being applied where they were not intended.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger phrases are broad and generic enough that the skill could activate for many loosely related requests involving browsers, automation, or productivity. In an agentic environment, over-broad activation can cause the wrong skill to run, leading to unintended workflow execution, confusing outputs, or unsafe automation being applied in the wrong context.

Vague Triggers

High

Confidence: 95% confidence
Finding: The skill’s activation scope is overly broad, using common terms like "browser," "automation," and generic workflow language that can match many unrelated user requests. This increases the chance of accidental invocation, causing the wrong skill to influence responses in contexts where its assumptions, workflows, or safety boundaries do not apply.

Vague Triggers

High

Confidence: 98% confidence
Finding: The keyword list contains vague, high-frequency terms such as "browser," "headless," "automation," and "cli," which are likely to collide with ordinary requests outside this skill’s intended scope. In an agent environment, this can lead to inappropriate routing or over-selection of the skill, reducing reliability and potentially surfacing unsafe or irrelevant workflow guidance.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger keywords are broad and include generic terms like browser, automation, cli, and bug fix, which can cause this skill to activate in many unrelated contexts. Over-broad activation increases the chance that the wrong skill handles a request, leading to confusing behavior, unsafe workflow suggestions, or bypass of more specialized controls in other skills.

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The description defines activation using a very wide set of user needs, including broad categories such as work-productivity, browser, headless, automation, and practical workflow support. Because the boundaries are not explicit, the skill may be selected for ambiguous requests outside its intended scope, which can reduce reliability and create unsafe or incorrect assistance paths.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The default prompt uses broad natural-language trigger wording tied to common terms like work-productivity, browser, automation, workflow, checklist, and analysis, which can cause the skill to be invoked in many unrelated conversations. Because implicit invocation is enabled, this increases the chance of unintentional activation, context leakage into the skill, and unsafe or confusing tool-routing behavior.

Vague Triggers

High

Confidence: 95% confidence
Finding: The trigger sentences are extremely broad and partially malformed, matching generic phrases like 'Help me' and 'I need a practical workflow' that commonly occur in ordinary user requests. This can cause unintended activation of the skill outside its intended scope, leading to prompt-routing errors, over-collection of context, or execution of browser/automation-oriented workflows when the user did not explicitly request them.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal