Back to skill

Security audit

Professional Web Engineering

Security checks for vulnerabilities and agentic risk

Overview

This is a broad but coherent web and app engineering guidance skill with no embedded code, hidden execution, persistence, or exfiltration behavior.

Install this if you want broad professional web/app engineering guidance. Because it covers deployment, databases, secrets, and production checks, review and explicitly approve any real-world changes to production systems, credentials, or data before letting an agent act.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill contains imperative language directives in Indonesian ('Gunakan skill ini ketika', 'Jangan gunakan untuk') and generally frames operation in Indonesian without any user opt-in or locale negotiation. Forcing a language or interaction mode can override user preferences, reduce transparency, and create policy/compliance issues in multilingual environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The activation guidance is overly broad and circular: 'Use when the conditions in the description are met' does not define concrete triggers or boundaries for when the skill should be invoked. In an agent setting, ambiguous activation scope can cause over-selection of this skill across unrelated tasks, increasing the chance that powerful engineering guidance is applied in unintended contexts and weakening policy-based routing controls.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description is extremely broad and covers nearly any web or app development task, which can cause the agent to activate in situations where a narrower, safer, or more specialized skill should be used. Overbroad activation increases the chance of inappropriate scope expansion, confused tool use, and bypass of user intent boundaries.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 1246)May include surrounding context.

md
- Desktop-only design
- No performance checks
- Known vulnerabilities unpatched
- Deploying without verification

## Rationalization Prevention