Back to skill

Security audit

text-to-image

Security checks for vulnerabilities and agentic risk

Overview

The scanners label this skill benign because its reported image-generation and related workflows appear purpose-aligned, while flagging privacy disclosures as areas to consider.

Before installing, consider whether prompts may contain confidential material and whether you are comfortable with external services processing them. Check whether you want generated images shared with downstream actors or scheduled for ongoing processing. The verdict is benign, but confidence is medium and Skillspector’s review is partial: three local references were unresolved, so the evidence is incomplete.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (2)

The generation workflow sends user prompts to external services without a privacy disclosure.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow sends user prompts, potentially including confidential material from spreadsheets or content calendars, to external image-generation services without a privacy review or explicit disclosure of the data flow. Although the providers are named and remote processing is expected for this functionality, automatic provider selection and execution of inexpensive runs can transmit sensitive content before the user approves that disclosure.

Content

No source excerpt is available for this finding.

The chaining workflow shares image datasets with downstream Actors without a privacy warning.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The chaining instructions pass a dataset reference to downstream Actors and recommend ongoing scheduled processing without a privacy or authorization checkpoint. This can make generated images and potentially accompanying prompts or metadata accessible to additional processing components; authorization to generate images does not necessarily authorize those subsequent uses.

Content

No source excerpt is available for this finding.