ClawHub
Back to skill

Security audit

Location Aware Backgrounds

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a straightforward image-generation workflow that saves generated background images and does not show hidden access, destructive behavior, or persistence beyond image files.

Before installing, make sure you are comfortable providing a Gemini API key and using Nano Banana Pro as the renderer. Generated images are saved locally, and any prompt details or reference images you provide may be used by the image-generation provider to complete the render.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The instruction says the skill 'always renders through $nano-banana-pro' and 'Do not switch to other image generators,' which imposes a fixed tool choice regardless of user preference. Under the policy scope here, forcing a specific language/locale or similar user-facing constraint without opt-in is disallowed; this is analogous to an imposed user-facing constraint with no choice offered.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
70% 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 35)May include surrounding context.

md
- asked for a live lookup or current-context lookup
   - asked for a location-aware result and has not opted out of live context

   Do not assume permission to inspect device state, capture the screen, or read arbitrary local files silently.

3. Resolve the output contract.
   Decide:

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

This markdown skill description states that if no output path is provided, it will save a timestamped PNG under ./generated/, which is a file-write operation affecting the user's filesystem. The document does not clearly warn the user about this default persistence behavior or call it out as a side effect before execution.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.