Back to skill

Security audit

Architecture Decision Record

Security checks for vulnerabilities and agentic risk

Overview

This skill only helps draft Architecture Decision Records and does not request code execution, credentials, persistence, or broad data access.

Installers should expect this skill to guide the agent in asking for ADR details and drafting a decision record. Be aware it may activate on some general decision-writeup requests, but its instructions remain limited to producing documentation.

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 (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The usage example "Help me write up why we chose [option] over [alternative]" is a broad, natural phrase that could arise in ordinary conversation outside a clear ADR context. Because the skill does not provide exclusion conditions or tighter invocation constraints, this example increases the risk of unintended activation.

Content

No source excerpt is available for this finding.