ClawHub
Back to skill

Security audit

openai-vision

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent OpenAI vision helper whose local-file and batch examples are purpose-aligned, though users should treat uploaded images and saved analysis files as potentially sensitive.

Install only if you are comfortable sending selected images to OpenAI for analysis. Do not process private documents, IDs, screenshots, regulated data, or confidential business images unless your organization allows it, and store batch output files as 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill includes examples that read local image files, base64-encode their contents, and send them to an external OpenAI API, but it does not prominently warn users that local file contents may contain sensitive information and will leave the local environment. In a vision-analysis skill this behavior is expected, but the lack of explicit disclosure increases the risk of unintentional data exfiltration of private documents, screenshots, IDs, or other sensitive imagery.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The batch-processing example writes analysis results to an output file without warning that the file may contain extracted OCR text, descriptions of sensitive images, or error details derived from private inputs. This can create an unintended local persistence channel for sensitive content, especially when processing directories of personal, regulated, or confidential images.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.