Error Message Improver

• Outcome: Pass
• Latest audit: Jun 14, 2026 at 8:44 AM UTC
• Version:
  0.1.0

Overview

This is a text-only skill for improving software error messages, with the main caveat that its trigger terms are overly broad and could make it activate too often.

Safe to install if you want help drafting or reviewing software error messages. Be aware that it may activate on generic requests containing words like better or error; explicit invocation is preferable until the trigger wording is tightened.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (10)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger phrase is broad enough to overlap with ordinary help requests, which can cause the skill to activate outside its intended scope. In an agent system, this increases the chance of prompt routing confusion, unintended invocation, and potential exposure to adversarial content embedded in otherwise normal user requests.

Vague Triggers

Medium

• Confidence: 93% confidence
• Finding: The trigger phrases are extremely generic help-seeking utterances and can cause the skill to activate in situations far outside its intended scope. This increases the chance of prompt hijacking of unrelated user requests, unexpected workflow injection, and user confusion about why this skill was invoked.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The activation keywords and usage guidance rely on broad terms like 'general-help', 'better', 'messages', and 'type', which do not meaningfully constrain when the skill should run. In an agent environment, ambiguous activation scope can make the skill fire on unrelated conversations, causing unintended behavior and making it easier for a low-trust skill to influence outputs outside its proper domain.

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The skill description and trigger terms are broad enough to match ordinary conversation about being 'better', 'help', 'error', or 'messages', which can cause the skill to activate outside its intended scope. Over-broad activation is dangerous because it can hijack unrelated user requests, inject irrelevant workflow steps, and increase the chance that a less appropriate skill handles sensitive or important tasks.

Vague Triggers

High

• Confidence: 98% confidence
• Finding: The Trigger section enumerates vague standalone keywords like 'better', 'error', 'messages', 'level', and 'type' without scope constraints, making accidental invocation highly likely in normal dialogue. In an agent system, this broad matching can distort routing decisions, causing this skill to preempt more suitable skills or respond in contexts where it lacks the necessary domain fit.

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The skill metadata description includes broad English trigger terms such as 'general-help', 'better', 'error', 'messages', and 'labels', which are common in ordinary user requests and can cause the skill to activate outside its intended scope. Over-broad activation increases the chance that the agent routes unrelated prompts through this skill, producing incorrect task selection, unintended instruction influence, or response confusion.

Vague Triggers

Medium

• Confidence: 97% confidence
• Finding: The explicit keyword list is highly unspecific and lacks contextual guards, so routine words like 'better', 'error', 'messages', 'level', 'beginner', or 'type' may spuriously invoke the skill. In an agentic system, this can degrade routing integrity and allow this skill's instructions to override more appropriate skills for unrelated tasks.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The example trigger phrase 'Help me Better Error Messages.' is vague and unnatural, giving weak guidance for reliable scoped activation and making it harder to distinguish legitimate use from incidental language overlap. Poor examples reinforce over-broad matching behavior and contribute to accidental invocation of the skill.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The manifest enables implicit invocation with a very broad prompt/description around generic terms like 'help', 'better', 'error', and 'messages', without clear scoping constraints. This can cause the skill to be auto-selected in unintended contexts, increasing the chance of prompt-surface expansion, irrelevant activation, or misuse in conversations that only loosely match the topic.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger sentences are broad enough to match ordinary user requests like asking for help with better error messages, which can cause this skill to activate outside a narrow, explicitly intended scope. Over-broad invocation increases the chance of routing unrelated conversations into this skill, leading to inappropriate automation, confused handling, or interference with more suitable skills.

VirusTotal

63/63 vendors flagged this skill as clean.