Back to skill

Security audit

GPT Image 2 种草图片

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly an AI Hive image-generation helper, but it needs review because its packaged script has broader unused capabilities and under-scoped credential, network, and file-download handling.

Install only if you trust AI Hive and the skill publisher, keep the default HTTPS base URL unless you intentionally operate a trusted endpoint, avoid running it with elevated privileges, and be aware that selected media files and prompts are sent to the configured service while downloaded results are written locally.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (4)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/imagegen.py:399
Finding

Path Traversal Through an Untrusted API-Supplied Output Filename

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/imagegen.py:91
Finding

Bearer API Key Can Be Transmitted to an Arbitrary Configured Endpoint

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/imagegen.py:281
Finding

Unvalidated Remote Upload and Download URLs Enable Data Exfiltration and SSRF

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
SKILL.md:72
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (11)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill is presented as a narrowly scoped GPT Image 2 image-generation helper, but the detected behavior extends to generic chat, video generation, account and wallet queries, media upload, browser-based API-key setup, and local config storage. This mismatch is dangerous because users may grant trust and permissions based on the benign description while the underlying implementation can access unrelated sensitive functionality and potentially expose credentials or account data.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This markdown file contains user-facing natural-language content exclusively in Chinese, and there is no indication that the skill is region-specific or that users can opt into another language. Under the policy rule, forcing a specific language without user choice is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill invokes local Python scripts and installation commands but does not declare any explicit tool or permission scope, leaving shell, filesystem, environment, and network access implicit. That increases the chance that an agent or reviewer will authorize broader capabilities than the skill’s stated purpose requires, enabling unintended command execution, file access, or outbound requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The description and instructions are written as direct Chinese-language requirements for generating content, with no indication that users may choose another language or locale. This can violate language/locale policy when a skill implicitly mandates a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file is presented as a narrowly scoped GPT Image 2 seeding-image skill, but the implementation and module docstring describe and retain a general-purpose AI Hive client. This creates a capability mismatch that can mislead reviewers and users, increasing the risk that broader functions are invoked or repurposed outside the declared skill boundary.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The top-level documentation explicitly describes a generic AI capability tool, which contradicts the declared identity of a GPT Image 2 seeding-image skill. This kind of scope ambiguity is dangerous because trust, approval, and deployment decisions may be made based on the metadata while the codebase actually contains broader capabilities.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This code embeds the skill’s primary description and user-facing CLI guidance in Chinese only, which imposes a specific language on users without opt-in. The policy for natural-language violations applies to code files when string literals or comments enforce a locale choice.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Embedding video-generation capability in a skill advertised only for GPT Image 2 seeding images violates least functionality and expands the attack surface unnecessarily. Even if not exposed through the final skill entrypoint, keeping unrelated media-generation code in the artifact raises the chance of accidental exposure, misuse, or future wiring into the skill without proper review.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill-specific CLI exposed to users uses Chinese-only descriptions and help strings for commands and arguments, with no alternative locale or opt-in. That constitutes a language policy violation because the skill fixes the interaction language rather than letting the user choose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The example API key placeholder uses Chinese text ("请替换为你的APIKey"), which imposes a specific language in user-facing configuration content. There is no indication that the skill is region-specific or that other language/localization options are supported.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

Text chat, model enumeration, and user-info inspection are broader account and platform capabilities than required for a fixed-purpose image skill. While not inherently dangerous on their own, they increase privilege and information exposure beyond the stated use case, enabling unnecessary discovery of available models and account metadata.

Content

No source excerpt is available for this finding.