Security audit

Software Data Admapix Raw Developer Helper

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only workflow helper with overly broad auto-invocation wording but no hidden execution, credential handling, persistence, or destructive behavior.

Before installing, be aware that this skill may activate for generic software or data prompts because its triggers are broad. It is suitable as a lightweight workflow/checklist helper, but users may want narrower triggers if they only intend it for AdMapix-specific raw-data workflows.

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 (12)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The trigger sentence is excessively broad and resembles ordinary conversational text, which can cause the skill to activate in contexts the user did not intend. In an agent ecosystem, ambiguous activation increases the chance of prompt routing errors, unintended execution paths, and misuse of the skill for unrelated requests.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: Although additional words are present, the trigger remains ambiguous because it is still phrased as a generic request for a 'practical workflow' rather than a precise capability boundary. This makes accidental or overbroad matching more likely, which can cause the wrong skill to handle requests and weaken safety controls that depend on accurate routing.

Vague Triggers

High

Confidence: 97% confidence
Finding: The trigger phrases are broad and generic enough to match ordinary help requests, which can cause the skill to activate outside its intended AdMapix/raw-data workflow. In an agent ecosystem, overbroad routing can divert unrelated user prompts into this skill, increasing the chance of inappropriate actions, incorrect guidance, or unintended access to adjacent workflows.

Vague Triggers

High

Confidence: 96% confidence
Finding: The skill description is broad and maps to generic user intents like 'software-and-data' and 'implementation support,' which can cause the skill to activate for many unrelated requests. Over-broad invocation increases the chance of prompt-routing mistakes, where users receive irrelevant or overly trusted workflow guidance in contexts the skill was not designed for.

Vague Triggers

High

Confidence: 98% confidence
Finding: The trigger keywords include highly generic terms like 'raw,' 'data,' 'layer,' 'apps,' and 'bug fix,' which are common across many unrelated requests. This makes accidental invocation likely, potentially hijacking routing decisions and causing the wrong skill to handle sensitive engineering or data tasks.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The example trigger sentences use vague language and broad help-seeking phrases instead of well-bounded examples tied to a precise task. This encourages loose matching behavior and reinforces accidental activation on generic requests, reducing routing reliability and increasing the chance of inappropriate skill use.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The trigger keywords include very generic terms such as raw, data, layer, apps, and bug fix, which are common across unrelated requests. This can cause the skill to activate outside its intended scope, leading to incorrect routing, prompt interference, or accidental application of domain-specific guidance in unrelated contexts.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill description uses broad natural-language activation criteria like handling requests involving software-and-data, admapix, raw, data, or layer without clear boundaries. Ambiguous invocation rules increase the chance of over-triggering and make it harder for the orchestrator or user to distinguish this skill from more appropriate ones.

Vague Triggers

Low

Confidence: 85% confidence
Finding: The example trigger phrases are written as generic help requests and do not include disambiguating constraints or exclusions. Because they resemble ordinary user language, they may cause the skill to match routine requests that are only loosely related, increasing accidental invocation risk.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The default prompt is written as a broad natural-language trigger tied to common terms like software, data, workflow, checklist, analysis, and implementation support. Because implicit invocation is enabled, this increases the chance the skill is auto-selected for unrelated user requests, which can cause prompt-scope confusion, unintended execution paths, or exposure of the skill in contexts the user did not explicitly request. The broad demand-oriented wording in this skill makes the issue more dangerous, not less, because it overlaps heavily with normal developer-assistance queries.

Vague Triggers

High

Confidence: 97% confidence
Finding: The trigger sentence begins with a very broad everyday phrase ('Help me') and then appends requirement text, making activation conditions overly permissive and likely to match unrelated user requests. In an agent skill-routing context, this can cause accidental invocation, misrouting, and inappropriate use of the skill in contexts the user did not intend.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: Although the second trigger adds context, it still remains ambiguous because it is phrased as a generic request for 'a practical workflow' followed by long requirement text rather than a precise, user-facing intent. This increases the chance of false-positive routing, where unrelated workflow or software requests are captured by this skill and lead to confusing or unsafe task execution paths.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.