Back to skill

Security audit

Plugin Intelligence

Security checks for vulnerabilities and agentic risk

Overview

This skill is a guidance-only plugin orchestration skill with no executable payload, and its higher-risk concepts are mostly framed with validation, permission checks, and least-privilege cautions.

Before installing, confirm you are comfortable with a skill that influences how the agent chooses and coordinates plugins. It does not ship code or request credentials, but because plugin orchestration can lead to write, send, delete, or admin actions through other plugins, users should keep approving sensitive plugin actions explicitly and ensure plugin permissions are checked for each task.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The usage guidance is overly broad and permissive, encouraging activation for vague 'conditions in the description' and broad 'high-level plugin reasoning' tasks without clear boundaries or preconditions. In a plugin-orchestration skill, ambiguous triggers can cause the agent to invoke the skill in inappropriate contexts, increasing the chance of unsafe plugin selection, unnecessary permission use, or risky multi-plugin operations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill switches into Indonesian directive language for core usage guidance ('Gunakan skill ini ketika', 'Jangan gunakan untuk'), and this pattern continues throughout the document as mandatory operating instructions. There is no indication that language selection is optional or user-driven, which creates a language/locale policy concern for users expecting another language.

Content

No source excerpt is available for this finding.