Back to skill

Security audit

AI鞋履试穿效果图|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a disclosed AI-HIVE footwear image workflow, but it has review-worthy security issues around API-key handling and downloaded file writes.

Review before installing. Use only with a dedicated AI-HIVE key, do not set `--base-url` or `AI_HIVE_BASE_URL` unless you fully trust the endpoint, keep generated downloads in a disposable output directory, and confirm all billable generation parameters before running. Consider rotating any API key previously used with an untrusted base URL.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/imagegen.py:91
Finding

Bearer API Key Disclosure Through an Unrestricted Base URL Override

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/imagegen.py:399
Finding

Path Traversal and Arbitrary File Overwrite Through Server-Controlled Download Names

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:67
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (10)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The declared description promises a full AI鞋履试穿 capability and explicitly mentions operational API behaviors for generating assets. The supplied code does not perform any of those core functions. It only creates a static production brief in JSON from command-line arguments. While the brief may be a supporting preproduction artifact related to the domain, the primary behavior is materially narrower and different from the declared skill purpose, so this is a mismatch.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill includes executable code examples that use environment variables, local files, shell commands, and network access, but it does not declare any tool scope or permissions boundaries. This creates an authorization gap where an agent or reviewer may not clearly understand what capabilities the skill expects, increasing the risk of over-privileged execution, accidental secret exposure, or unintended external calls.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The default prompt hard-codes Chinese output without checking the user's language preference or obtaining opt-in. While not a direct code-execution issue, this can degrade user understanding of generated workflow steps and runnable commands, increasing the risk of user error, mistaken approval, or unsafe execution of API-related actions in a language they did not request.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest enables implicit invocation without any visible trigger scoping, narrowing conditions, or safety constraints. In a skill that can generate runnable AI-HIVE commands and initiate media-generation workflows, broad auto-invocation increases the chance the agent activates in unintended contexts, potentially causing unauthorized API actions, unnecessary cost, or processing of sensitive/unauthorized images.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The file header and implementation clearly indicate this script is a general-purpose AI Hive OpenAPI client supporting chat, image, video, model enumeration, upload, and task polling, while the manifest presents it as a narrow footwear try-on preview skill. This capability overreach violates least privilege for skills and can let a caller use the skill as a broader AI gateway than intended, expanding abuse surface and policy bypass opportunities.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file’s natural-language interface, descriptions, prompts, and user-facing messages are written entirely in Chinese, including the module docstring and command help text. There is no indication that the skill is region-specific or that users may opt into another language, which can violate language/locale policy for general-purpose skills.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Text chat is a broad, open-ended capability unrelated to a constrained footwear try-on preview skill. It can be repurposed for arbitrary model interaction, prompt handling, and multimodal input processing outside the declared business scope, making the skill a covert general AI access path.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Generic video generation falls outside the stated purpose of a footwear try-on image-preview skill and materially broadens what users can do through the integration. This increases operational risk, content-policy evasion opportunities, and cost exposure by enabling a different media modality than the manifest promises.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The docstring, argparse description, and embedded skill name are all presented only in Chinese, with no indication that users can select another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill exposes a user-info endpoint that returns account and wallet-related information not described in the footwear try-on manifest. Even if low impact, this unnecessarily grants access to account metadata and billing context unrelated to the skill's stated purpose, increasing information exposure and reconnaissance value.

Content

No source excerpt is available for this finding.