Security audit

Error Message Improver

Security checks for vulnerabilities and agentic risk

Overview

This skill is a network troubleshooting helper published under an error-message name, with broad automatic triggers, so it needs review before installation.

Install only if you want a network-diagnostics helper, not an error-message improver. The publisher should rename it to match its real purpose, narrow the trigger conditions to explicit network troubleshooting requests, and reconsider implicit invocation.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (16)

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The README claims the skill is an 'Error Message Improver' but the actual content describes a network-diagnostics workflow for bufferbloat, routers, modems, and latency. This capability mismatch can cause the agent or user to invoke the skill in unintended contexts, leading to over-broad behavior, misrouting of requests, and unsafe trust in a skill whose documented purpose does not match its actual operational scope.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The skill’s declared name and title suggest an error-message improvement function, but the manifest and body actually define a network-diagnostics workflow. This mismatch can cause incorrect routing, unsafe invocation in unrelated contexts, and operator confusion about the skill’s true purpose, which is especially risky in agent systems that rely on metadata for selection and trust decisions.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The manifest description contains broad activation terms such as general-help and common networking words, which can match many ordinary user requests outside the intended scope. In an agent environment, overbroad matching can trigger this skill unexpectedly, leading to inappropriate guidance, misrouting, or unintended interference with other skills.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill is named and presented as an error-message improver, but the body actually targets network troubleshooting workflows. This mismatch can cause the wrong skill to be invoked or trusted under false pretenses, leading to incorrect task routing, confusing outputs, and reduced operator ability to review what the skill really does.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger list includes broad terms such as general-help, router, modem, and latency, which commonly appear in benign conversation. Overbroad triggers increase the chance of accidental activation, causing the skill to intercept unrelated requests and potentially override more appropriate skills or workflows.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger phrase 'Help me ...' is so broad that it can match ordinary conversation and cause unintended skill activation. In an agentic environment, over-broad activation can misroute user requests, override more appropriate skills, and increase the chance of unsafe or irrelevant actions being taken under the wrong context.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The README documents a network-diagnostic workflow while the skill is named 'error-message-improver', creating a material mismatch between the advertised capability and actual behavior. This can mislead routing, review, and user trust, making it easier for a skill to be invoked in unintended contexts or to evade scrutiny by appearing to serve a different purpose.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrases are broad and generic enough to match common help requests, increasing the chance of unintended activation. In combination with the capability mismatch, this broad routing surface could cause the skill to intercept unrelated prompts and influence outputs outside its intended domain.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The trigger phrases are broad and permissive, including generic terms like 'general-help' and open-ended prompts that can activate the skill for loosely related requests. In an agent environment, weak activation constraints increase the chance of accidental invocation, prompt-routing abuse, or the skill handling tasks outside its intended safety-reviewed scope.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The keyword list includes generic terms that lack contextual constraints, increasing the chance of accidental activation on unrelated conversations. While this does not directly execute code, it weakens skill-boundary enforcement and can degrade safety by causing the wrong capability to engage.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: Using a common phrase like 'Help me' in an example trigger normalizes extremely vague activation patterns and may encourage downstream systems or authors to treat nonspecific language as sufficient for invocation. In context, this makes the skill more likely to activate on benign, unrelated requests and undermines safe intent discrimination.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The trigger section provides only generic keywords and example phrases without defining scope boundaries or exclusion conditions. In a skill-routing system, that ambiguity makes unintended invocation more likely and reduces confidence that the skill will activate only for suitable network-diagnostics tasks.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The skill metadata is internally inconsistent: the name suggests an error-message utility, while the description and default prompt target broad network-diagnostics assistance. Combined with allow_implicit_invocation=true, this creates an overly broad and misleading invocation scope that could cause the agent to call the skill in unintended contexts. That increases the chance of inappropriate delegation, confused-user interactions, and misuse of the skill beyond its intended boundaries.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The manifest identifies the skill as 'error-message-improver', which suggests functionality around rewriting or clarifying error messages. However, the documented requirement, usage signals, and implementation plan throughout this file are focused on local diagnostic workflows for network issues such as bufferbloat, router problems, modem issues, and latency, which is a materially different purpose.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The activation guidance is ambiguous because it mixes generic category labels with loose trigger sentences and does not define required conditions for invocation. This makes the skill easier to activate outside its intended scope, increasing prompt-routing errors and creating opportunities for an attacker or noisy input to steer the agent into the wrong workflow.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 75% confidence
Finding: The file is presented as a Chinese README but substantial user-facing instructions, including demand and usage text, remain in English. This creates an implicit language behavior without giving users a clear language or locale choice, which may conflict with language/locale policy expectations.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.