Boston

• Outcome: Pass
• Latest audit: Sep 10, 2026 · 9:08 PM UTC
• Version: 1.0.0

Overview

This is a static Boston city-guide skill with some high-stakes informational sections that users should verify, but it does not request dangerous access or perform hidden actions.

Install only if you want a static Boston reference. Treat medical, legal, immigration, tax, alcohol, transit, and pricing details as starting points, not authoritative current advice; verify important decisions with official sources or qualified professionals.

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

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

Findings (13)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The trigger text says the skill should be used when a user asks about Boston 'for any purpose,' which is extremely broad and lacks clear boundaries on when the skill should or should not activate. It does not provide specific trigger phrases, scope constraints, or negative examples to prevent accidental invocation during ordinary conversation about Boston.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: This file provides detailed business, tax, licensing, employment, and legal guidance with specific costs, rates, and procedural steps, but it does not warn that these requirements may change or that users should verify them with official Massachusetts agencies or qualified professionals. Because users may rely on outdated or incomplete legal/compliance information when forming a business, hiring staff, or filing taxes, the omission can lead to regulatory noncompliance, financial penalties, or legal disputes.

Intent-Code Divergence

Medium

• Confidence: 91% confidence
• Finding: The file states Massachusetts bans happy-hour drink specials, then later says many places do bottomless mimosas/bloody marys. That inconsistency can mislead users about alcohol laws and promotions, potentially encouraging them to seek or expect offers that may be illegal or unavailable. In a travel/local guidance skill, inaccurate legal guidance is more dangerous because users may rely on it for real-world behavior.

Natural-Language Policy Violations

Medium

• Confidence: 84% confidence
• Finding: This finding reflects overbroad or inaccurate locale-specific alcohol policy guidance at the same line, where the text implies bottomless alcoholic brunch promotions are typical. Misstating regulated alcohol rules can cause users to make decisions based on false legal assumptions, leading to compliance issues or unsafe planning. Because the skill is framed as practical city advice, users are likely to trust and act on it.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: This section gives actionable medical triage and care-navigation guidance, including when to use urgent care versus the ER, when to call 911, and how to seek mental health crisis support, but it does not clearly state that the content is general information and not a substitute for professional medical advice. Users may rely on simplified guidance in urgent situations, potentially delaying appropriate care or misjudging symptom severity.

Natural-Language Policy Violations

Medium

• Confidence: 82% confidence
• Finding: The 'Who Lives Here' section describes residents using generalized class-based labels such as 'older wealthy residents,' 'international students with money,' and 'established couples without kids.' These statements present demographic and socioeconomic generalizations as factual neighborhood characterizations without context or qualification, which is a natural-language policy concern in user-facing content.

Natural-Language Policy Violations

Medium

• Confidence: 84% confidence
• Finding: Phrases like 'Old money Bostonians,' 'Politicians,' and 'Young professionals wanting quiet elegance' characterize who belongs in the neighborhood using socially loaded identity and class cues. This can introduce biased or exclusionary framing in a descriptive skill file and is a natural-language quality and policy issue.

File System Enumeration

Medium

• Category: Data Exfiltration

Content

250,000 students, city is used to them.
- Campus police in addition to BPD
- Most crime = petty theft
- Don't walk home drunk alone

---

• Confidence: 60% confidence
• Finding: Code scans file system directories looking for sensitive files. This could be reconnaissance for credential theft.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The document provides detailed immigration pathways, timelines, eligibility summaries, and procedural steps without a prominent upfront disclaimer that the information may become outdated and is not legal advice. Because immigration rules, filing windows, and travel/work authorization consequences can change frequently, users may rely on stale or incomplete guidance and make decisions that risk status violations, missed deadlines, or improper filings.

Natural-Language Policy Violations

Medium

• Confidence: 94% confidence
• Finding: The text normalizes and lightly instructs unlawful and unsafe street-crossing behavior by saying 'Everyone does it' and advising readers to 'Follow local traffic rhythm.' In a visitor guidance skill, that can encourage tourists unfamiliar with Boston traffic patterns to take risks, increasing the chance of injury or citation.

Natural-Language Policy Violations

Low

• Confidence: 81% confidence
• Finding: The phrase "Language barriers sometimes" embeds a language-related constraint in the content and can discourage or disadvantage users based on language expectations. Under the policy, language/locale constraints should be optional, justified, or framed with user choice rather than presented as a blanket characteristic.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: SQP-3 applies to all file types and covers language or locale policy concerns in natural-language content. Line L075 highlights an English-language limitation for staff but does not provide any opt-in, accommodation guidance, or justification, which can be interpreted as a language-policy concern.

Missing User Warnings

Low

• Confidence: 88% confidence
• Finding: The content references 'after-hours clubs' and notes they are 'rare, not legal,' but still presents them as a late-night option in a lifestyle guide. In a city-navigation skill, even brief mention of illegal venues can normalize unlawful activity or be interpreted as a recommendation, especially without an explicit safety or legal caution.

Static analysis

No suspicious patterns detected.