Bucharest

• Outcome: Review
• Latest audit: Sep 10, 2026 · 10:53 PM UTC
• Version: 1.0.0

Overview

This is a non-executing Bucharest guide, but its visa workaround guidance is risky enough that users should review it before installing.

Treat this skill as a general Bucharest reference only. Before relying on it for immigration, work authorization, taxes, healthcare, or severe dietary needs, verify with official authorities or qualified professionals; in particular, do not treat the tourist visa hopping or enforcement comments as legal advice.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (15)

Intent-Code Divergence

High

• Confidence: 95% confidence
• Finding: The skill presents 'Tourist Visa Hopping' as 'Legal but not ideal long-term' even though the document elsewhere states tourist stays are limited to 90 days within a 180-day period. That contradiction can mislead users into overstaying or attempting border-run behavior that may violate Romanian/EU immigration rules, exposing them to fines, entry bans, or other legal consequences.

Vague Triggers

Medium

• Confidence: 96% confidence
• Finding: The activation condition says the skill should be used whenever a user asks about Bucharest 'for any purpose,' which is overly broad and can cause the agent to invoke this skill outside its intended travel/relocation guidance scope. Overbroad routing can lead to inappropriate responses, irrelevant file loading, or the skill being preferred over safer or more specialized skills, though this file contains no direct code execution or prompt-injection payloads.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The dietary guidance discusses gluten-free, kosher, and halal options using uncertain language such as 'ask', 'often not certified', and 'always ask' without a prominent safety warning that restaurant claims may be unreliable and that severe allergies, celiac disease, or strict religious requirements require independent verification. In a practical city guide, users may rely on this advice for real-world food decisions, creating a plausible risk of allergic exposure, cross-contamination, or accidental consumption of non-compliant food.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: This section discusses immigration 'workarounds,' variable enforcement, and practical ways to remain in Romania without a strong, prominent disclaimer that laws change and only official/legal sources should be relied on. In a relocation/visa guidance skill, users are likely to treat this as actionable compliance advice, increasing the risk of unlawful stay, unauthorized work, or bad-faith reliance on outdated information.

Natural-Language Policy Violations

Low

• Confidence: 84% confidence
• Finding: Line L235 states 'Romanian language' as a blanket requirement in natural language. As written, this imposes a locale/language constraint without clarifying that it is a legal compliance requirement or offering context, which can violate the language/locale policy for skill content.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: Line L071 states that government offices are 'often Romanian only,' which is a locale constraint presented as a blanket condition without user choice or contextual framing. While this is informational travel guidance rather than an operational instruction, it still reads as a language constraint and may conflict with a policy preference to avoid imposing a specific language without opt-in.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: Line L238 states 'Romanian language helps,' which introduces a locale-specific recommendation in natural language. While mild and likely practical, it does not offer an opt-in or alternative for users who may not speak Romanian.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: The document is explicitly scoped to Bucharest and Romanian cuisine, which imposes a specific regional/localized focus without offering any user choice or opt-in. Under the stated policy rubric, locale constraints should be user-selectable or clearly justified as region-specific; the title presents the constraint directly but does not explain that limitation.

Natural-Language Policy Violations

Low

• Confidence: 87% confidence
• Finding: Line L161 says 'Romanian language' under the Public healthcare section, which presents a language limitation as a blanket characteristic. This can be read as a locale/language policy constraint without opt-in or nuance about availability of other language support.

Natural-Language Policy Violations

Low

• Confidence: 81% confidence
• Finding: The healthcare guidance highlights providers as 'good English' and the comparison table states 'English OK' for the expat belt, which implicitly centers English-language use as the expected default. Because the file does not offer any language preference choice or clarify that this is tailored specifically for English-speaking users, it can be read as a locale/language policy issue.

Natural-Language Policy Violations

Low

• Confidence: 76% confidence
• Finding: Line L330 instructs users to 'Bring translator for serious matters,' and the surrounding lines emphasize English availability as the relevant language. This implicitly centers a specific language without user opt-in or acknowledging other language needs, which can conflict with language/locale policy expectations.

Natural-Language Policy Violations

Low

• Confidence: 89% confidence
• Finding: Line L266 states 'English as primary language' as a general recommendation. This is a natural-language locale preference presented as a default, without offering a language choice or clearly limiting the guidance to a justified region-specific compliance need.

Natural-Language Policy Violations

Low

• Confidence: 90% confidence
• Finding: The instruction 'LinkedIn profile in English' prescribes a specific language rather than presenting it as an option based on user goals. Under the policy, forcing a language without user opt-in can be a natural-language policy violation, and this file does not offer alternatives or explain a mandatory locale requirement.

Natural-Language Policy Violations

Low

• Confidence: 82% confidence
• Finding: The communication section presents English as the default standard for meetings and documentation without giving users a choice or clarifying that this is only a general market tendency. This can be interpreted as a language-policy constraint not explicitly justified as mandatory.

Natural-Language Policy Violations

Low

• Confidence: 83% confidence
• Finding: Line L009 highlights that tours are available in English, and the overall guide is written for an English-speaking audience without any note about language alternatives or user preference. This can be a minor locale-policy issue when a skill presents a fixed language assumption rather than offering a choice or explaining that the content is intended for English-speaking travelers.

Static analysis

No suspicious patterns detected.