ClawHub
Back to skill

Security audit

Engineer

Security checks for vulnerabilities and agentic risk

Overview

This is an instruction-only engineering reasoning skill with optional local notes and no hidden execution, networking, credential use, or destructive behavior found.

Before installing, decide whether you want the optional ~/engineer/ memory. If enabled, keep it to non-sensitive preferences and engineering defaults, and avoid storing credentials, proprietary drawings, regulated data, or confidential process details unless you intentionally want local persistence.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

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 · failure-modes-first.md (reported line 35)May include surrounding context.

md
- corrective action: what permanently reduces recurrence
- verification: how to know the fix really worked

Closing an incident without verification is paperwork, not engineering.

## Quick Risk Matrix

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The file says the skill should activate for requests like specs, trade-offs, validation plans, or root cause analysis, and also asks to learn 'when this should activate later.' These are common categories of everyday technical help and the file does not provide narrowing constraints or negative examples, which could cause unintended invocation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.