Back to skill

Security audit

Educational Content & How-To

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed educational content-planning skill with accuracy guardrails and no executable install or hidden system behavior.

Installers should treat this as a content strategy aid, not an authority source. Review any public posts before publishing, verify cited marketing claims and topic facts, and use extra human review for health, finance, legal, safety, or other high-impact how-to content.

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

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The description says to use this skill when someone wants to "teach their audience, make a how-to / tutorial / step-by-step / explainer / myth-buster, build authority content, or earn saves," which spans several common marketing intents and does not clearly state exclusion conditions. Although the skill later distinguishes itself from siblings, the activation description itself remains broad enough to overlap with many ordinary content-generation requests.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The eval input 'Help me make an educational post teaching how to do X.' is very broad and can match ordinary user requests without clearly constraining scope, safety boundaries, or required expertise checks up front. In a routing or invocation context, this can cause the skill to be selected for underspecified or sensitive how-to topics, increasing the chance of unsafe educational output before YMYL or expertise guardrails are applied.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

The example 'Make us some how-to content.' is ambiguous and does not state topic boundaries, expertise limits, or whether the request is strategic planning versus full content generation. That ambiguity can lead to overbroad invocation of the skill and accidental use on topics that need stronger qualification, especially because this skill is designed to produce authoritative teaching content.

Content

No source excerpt is available for this finding.