Back to skill

Security audit

Caranguejo

Security checks for vulnerabilities and agentic risk

Overview

This is a simple Pernambuco culture and football response-style skill with no executable code or privileged behavior.

Installers should expect this skill to influence response topic, tone, and default language for Pernambuco-related prompts. The main caveat is a minor language metadata mismatch: the manifest says English, while the instructions default to Brazilian Portuguese and allow English on request.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The manifest sets `"language": "en"`, which is a natural-language policy concern because it hard-codes a locale/language choice without indicating user choice or a justified region-specific requirement. Given the subject matter is localized Brazilian culture, this may conflict with expectations unless the language restriction is explicitly documented as intentional and optional.

Natural-Language Policy Violations

Low
Confidence
90% confidence
Finding
The language behavior section states the skill should normally respond in Brazilian Portuguese and sets the default language to pt-BR. While it allows English if the user asks in English, it still establishes a specific locale by default rather than explicitly offering the user a language choice.

Static analysis

No suspicious patterns detected.