Miami

• Outcome: Pass
• Latest audit: Sep 10, 2026 · 8:46 PM UTC
• Version: 1.0.0

Overview

This is a markdown-only Miami guide with some broad or opinionated local guidance, but no evidence of hidden execution, persistence, credential access, or data exfiltration.

Installers should treat this as a static local guide, not an authoritative or always-current source. Verify prices, safety, transit, venue, and weather details before making plans, and use judgment around subjective neighborhood descriptions or language/culture assumptions.

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

• 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 (9)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The skill activates on essentially any Miami-related request, which is overly broad for a domain skill and can cause unintended invocation on generic queries where the user did not explicitly want this specialized behavior. While this is not a code-execution or data-exfiltration issue, it can degrade routing accuracy, overshadow other more appropriate skills, and increase the chance of misleading or overconfident location-specific guidance being injected into unrelated conversations.

Natural-Language Policy Violations

Medium

• Confidence: 89% confidence
• Finding: This markdown file contains natural-language guidance that elevates one language as preferred in a general way. Under the policy, language or locale constraints should be optional, justified, or framed as user choice rather than presented as a default expectation.

Natural-Language Policy Violations

Medium

• Confidence: 94% confidence
• Finding: The phrase 'Spanish essential' presents a language requirement as categorical guidance without user opt-in or nuance, which can steer users away from a neighborhood based on language ability and reinforce exclusionary assumptions. In a relocation/neighborhood recommendation skill, this is more sensitive because users may rely on it for housing decisions, amplifying potential discriminatory impact.

Natural-Language Policy Violations

Low

• Confidence: 78% confidence
• Finding: This markdown file contains natural-language descriptions such as 'Finance bros, influencers' that impose a particular cultural framing rather than neutral locale-aware guidance. Because SQP-3 applies to all file types and covers language/locale policy concerns, this can be flagged as potentially insensitive language without user opt-in or justification.

Natural-Language Policy Violations

Low

• Confidence: 76% confidence
• Finding: The 'Cuban Sandwich Rules' section uses absolute language such as 'Never' and 'Must,' which imposes a single normative standard rather than presenting it as a preference or common local convention. This can be read as a natural-language policy issue because it does not offer opt-in, alternatives, or contextual framing.

Natural-Language Policy Violations

Low

• Confidence: 82% confidence
• Finding: The "Best For" section explicitly lists "Jewish community members," which introduces religion-based audience targeting in natural-language guidance. This can be a policy concern because it frames recommendations around a protected characteristic without explaining a user-requested or region-specific need.

Natural-Language Policy Violations

Low

• Confidence: 76% confidence
• Finding: The heading "Young Professional (25-35, Single)" imposes a fixed age-based profile framing in natural language without indicating that these are optional examples or that users outside this range can still use the guidance. This can be interpreted as a policy-relevant demographic constraint presented without opt-in or alternative wording.

Natural-Language Policy Violations

Low

• Confidence: 77% confidence
• Finding: The description includes the phrase "Old money, established families," which imposes a specific socio-economic framing in the skill content. While not a language/locale restriction, it is natural-language guidance that may present a biased characterization as factual without qualification.

Natural-Language Policy Violations

Low

• Confidence: 87% confidence
• Finding: The content explicitly characterizes the neighborhood's international demographic using nationality-specific labels without an evident user need or opt-in. In a neighborhood guide, this can enable profiling, reinforce stereotypes, or create fairness and discrimination concerns, even though it is not a classic software exploit.

Static analysis

No suspicious patterns detected.