ClawHub
Back to skill

Security audit

Adversarial Alignment (Agent Smith)

Security checks for vulnerabilities and agentic risk

Overview

This is a small markdown-only plan-review skill that may activate on broad phrases but does not request tools, credentials, persistence, or execution authority.

Install only if you want an adversarial review style available during planning conversations. Be aware that generic phrases like "challenge plan" or "governance violation" may invoke it more often than expected, but the inspected package does not perform actions or access sensitive data.

A.I.G

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

SkillSpector

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
92% confidence
Finding

The trigger phrases are broad, generic action terms such as 'challenge plan', 'harden plan', 'block unsafe', and 'governance violation', which can plausibly appear in ordinary coordination or safety conversations. This increases the chance of unintended activation, causing the skill to inject adversarial behavior into unrelated interactions and potentially disrupt normal agent decision-making.

Content

No source excerpt is available for this finding.