Back to skill

Security audit

Brainstorming

Security checks for vulnerabilities and agentic risk

Overview

This is an instruction-only brainstorming skill that may add extra clarification steps, but it does not request sensitive access or install executable code.

Install this if you want the agent to pause and clarify ambiguous build requests before coding. Expect more questions and possibly Chinese workflow text unless your agent adapts language; review any downstream skill-compounding action before allowing persistent skill creation.

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 (3)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill auto-activates on very broad conditions such as any new feature request or vague 'help me build X' prompt, which can overlap with common benign user requests and unexpectedly take over the interaction flow. This creates a prompt-scope/control risk: the agent may invoke this skill without clear user intent, forcing extra questioning and downstream chained actions that can derail tasks or override user expectations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The entire skill is written as Chinese-only operational guidance, including trigger phrases and workflow text, with no indication that users may interact in another language or opt into Chinese. That creates a natural-language locale restriction without user choice or documented justification.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The command section repeats unconditional automatic activation for 'ambiguous requirements' without defining scope, conflict resolution, or concrete negative examples. This ambiguity increases the chance of unintended invocation and can amplify the impact of the earlier broad triggers by making activation behavior harder to constrain or audit.

Content

No source excerpt is available for this finding.