Back to skill

Security audit

xquads-hormozi-chief

Security checks for vulnerabilities and agentic risk

Overview

This is a Portuguese business-advice routing skill with broad triggers and some inconsistent role wording, but it contains no executable code, persistence, credential access, or hidden data handling.

Install only if you want a Portuguese-language Hormozi-style business coaching workflow. Be aware that it may activate for broad business terms and may give direct advisory content rather than only routing to another specialist.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The description says to use this skill for essentially any request about offers, leads, pricing, sales, and business scaling, which is overly broad. Overbroad activation can cause the skill to intercept generic business conversations, overshadow more appropriate or safer skills, and expand its influence beyond its intended niche.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The documentation explicitly forbids direct consulting, but later requires the skill to act as the specialist and deliver consulting output. This contradiction is dangerous because inconsistent instructions undermine policy enforcement, make behavior unpredictable, and can bypass governance or approval flows that depend on the skill remaining a non-executing router.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger condition 'Quando o usuário pedir ajuda com negócio' is ambiguous and broad enough to match many unrelated business-support prompts. In a multi-skill environment, this can lead to unintended invocation, prompt hijacking of benign requests, and inconsistent routing outcomes.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill states that it should only diagnose and route, but then instructs itself to change persona and perform the downstream specialist consulting directly. This creates a scope-expansion/instruction-conflict issue: callers and higher-level orchestrators may rely on the skill as a narrow router, while it actually executes substantive advisory behavior that should belong to separate skills with separate review boundaries.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructions and examples are written entirely in Portuguese and do not state that the user may choose another language. This can violate language or locale policy when the skill implicitly constrains interaction language without user opt-in or justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list contains broad, common business terms such as "business," "sales," "pricing," and "launch," which can match many unrelated user requests and cause the orchestrator to activate unexpectedly. Because this skill routes users to additional specialists, overbroad activation can lead to misrouting, prompt hijacking of otherwise general conversations, and expanded exposure of downstream skills beyond the user's intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The visible natural-language description is written in Portuguese, but the manifest does not indicate that the skill is region-specific or that users may choose their preferred language. This can create a language/locale policy issue if the broader environment expects language choice rather than a fixed locale.

Content

No source excerpt is available for this finding.