Security audit

USA Business Migration Planner

Security checks for vulnerabilities and agentic risk

Overview

The package appears to be a simple documentation skill, but its name says USA business migration while its actual instructions are for Dify/chat2dify workflow help, with broad implicit activation enabled.

Review this package carefully before installing because it may be invoked under a misleading business-migration name even though it actually provides generic Dify/chat2dify workflow assistance. It should be renamed or re-scoped and its implicit triggers narrowed before normal use.

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 (21)

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill metadata and README content are materially inconsistent: the skill is named 'USA Business Migration Planner' but the documented purpose is about Dify/chat2dify workflow support. This kind of mismatch can cause incorrect routing, unexpected activation, and user deception about what the skill actually does, which is dangerous in agent systems because it undermines trust boundaries and may trigger the wrong instruction set for unrelated requests.
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 identity and its actual behavior materially diverge: it is named as a USA business migration planner, but its workflow, examples, and triggers target Dify/chat2dify productivity tasks. This kind of semantic mismatch can cause the wrong skill to be invoked, mislead reviewers and users, and bypass intent-based routing or approval controls that rely on accurate metadata.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The top-level description says to use the skill for broad terms like work-productivity and for any request needing a workflow, checklist, analysis, or implementation support. Such generic activation criteria can cause accidental or excessive invocation, exposing users to irrelevant instructions and potentially overriding more appropriate, narrowly scoped 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 presents a major semantic mismatch: its name/body claim to be a USA business migration planner, while the actual description, demand statement, and triggers target Dify/chat2dify workflow assistance. This can cause the agent to invoke the skill in the wrong context, route unrelated user requests into an unintended workflow, and produce misleading or unsafe outputs due to scope confusion.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The invocation criteria are vague and expansive, stating the skill should be used whenever a user needs a practical workflow, checklist, analysis, or implementation support for the requirement. Such ambiguity weakens routing precision and can cause the skill to capture requests outside its intended domain, increasing the likelihood of confused-deputy behavior and user misdirection.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger keywords include overly broad generic terms such as "share" and "workflow," which can match many benign conversations unrelated to this skill. Over-broad activation increases the chance of prompt hijacking at the routing layer, accidental invocation, and irrelevant instructions being injected into user sessions where they do not belong.
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 labeled as a USA business migration planner, but its description and default invocation text instead reference an unrelated 'chat2dify' workflow. This mismatch can cause the wrong skill to be invoked implicitly, misleading users and upstream agents about the tool’s true purpose and enabling prompt-routing errors or unintended task execution.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill metadata says this is a 'usa-business-migration-planner', but the requirement plan is actually about a Dify/chat2dify workflow. This mismatch can cause the wrong skill to be invoked or trusted under false pretenses, leading to unintended actions, user confusion, or policy bypass through mislabeled capability routing.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The invocation examples do not define when the skill should and should not activate, leaving activation boundaries ambiguous. This ambiguity is risky because agent orchestrators or users may invoke the skill in unintended contexts, especially given the already mismatched skill identity and broad wording.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrase is broad enough to overlap with normal user language and could activate the skill unintentionally. In an agent environment, overbroad activation expands the skill's execution surface, increasing the chance of irrelevant instruction injection, confusion, or the application of the wrong workflow to user requests.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The README’s title and branding claim the skill is a 'USA Business Migration Planner', but the actual content describes a Dify/chat2dify workflow assistant. This mismatch can mislead users, reviewers, and routing systems about the skill’s real purpose, undermining trust and creating an opportunity for deceptive deployment or policy evasion.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger phrases are broad and map to ordinary requests for help or workflows, which increases the chance the skill activates outside its intended scope. In a skill system, overbroad activation can cause prompt hijacking of unrelated conversations, unexpected tool behavior, or accidental disclosure of skill instructions into benign contexts.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The title explicitly labels the skill as a USA business migration planner while the surrounding content documents a different function entirely. Even if not overtly malicious, this contradiction increases the chance of operator confusion, incorrect invocation, and trust failures in systems that depend on human-readable labels for security review and routing.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The trigger keywords include generic terms like workflow, share, and dsl, which are too ambiguous to safely distinguish intended use. In agent ecosystems, broad keywords increase unintended activations, reduce predictability, and can be abused to steer tasks into the wrong skill context.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 72% confidence
Finding: 该文件整体以 zh-CN 语言环境提供说明，但未见任何允许用户选择回复语言或说明该语言限制的自然语言声明。按规则，若技能隐含或强制特定语言/locale 而没有用户 opt-in 或明确合理性说明，属于自然语言策略问题。
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The default prompt uses a generic activation phrase and broad task wording, while the policy allows implicit invocation. In combination with the misleading metadata, this increases the chance that the skill is auto-selected for loosely related requests, causing unintended execution, confused-deputy behavior, or disclosure of irrelevant outputs in the wrong context.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger sentence begins with the broad phrase 'Help me', which is common in ordinary conversation and can cause accidental invocation of this skill outside its intended scope. In agentic systems, overly broad routing phrases increase the chance of misfires, causing irrelevant skill activation and potentially exposing the system to prompt-routing confusion.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The user-facing description embeds Chinese text inside an otherwise English README, but it does not explain any language-selection behavior or offer a user opt-in. This can conflict with language/locale policy expectations when a skill appears to impose or assume a language without user choice.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 81% confidence
Finding: Despite being a zh-CN README, key descriptive and trigger text is written in English, including user-facing invocation phrasing elsewhere in the file. This can amount to an implicit language policy choice without opt-in, which may conflict with expectations for locale-appropriate skill documentation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 82% confidence
Finding: The requirement statement is presented in Chinese within an otherwise English document, but the file does not explain whether the skill is intended for Chinese-language inputs only or whether users can choose their preferred language. This creates a potential language/locale policy issue because the skill behavior appears tied to a specific language context without explicit opt-in.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The sample trigger sentences include Chinese-language content as part of the activation wording, but the document does not clarify whether users must invoke the skill in Chinese or whether other languages are supported. For natural-language policies, a language-specific interaction pattern should be clearly optional or explicitly justified.
Content: No source excerpt is available for this finding.